/public/home/xuanbaby/rocm3.9-python3.6.8-tf1.15/bin/python3 /public/home/xuanbaby/rocm3.9-python3.6.8-tf1.15/lib/python3.6/site-packages/absl/flags/_validators.py:359: UserWarning: Flag --model_dir has a non-None default value; therefore, mark_flag_as_required will pass even if flag is not specified in the command line! 'command line!' % flag_name) 2021-04-21 10:30:52.698447: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libamdhip64.so 2021-04-21 10:30:54.255795: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1734] Found device 0 with properties: pciBusID: 0000:04:00.0 name: Device 66a1 ROCm AMD GPU ISA: gfx906 coreClock: 1.7GHz coreCount: 64 deviceMemorySize: 15.98GiB deviceMemoryBandwidth: 953.67GiB/s 2021-04-21 10:30:54.256007: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1734] Found device 1 with properties: pciBusID: 0000:26:00.0 name: Device 66a1 ROCm AMD GPU ISA: gfx906 coreClock: 1.7GHz coreCount: 64 deviceMemorySize: 15.98GiB deviceMemoryBandwidth: 953.67GiB/s 2021-04-21 10:30:54.256128: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1734] Found device 2 with properties: pciBusID: 0000:43:00.0 name: Device 66a1 ROCm AMD GPU ISA: gfx906 coreClock: 1.7GHz coreCount: 64 deviceMemorySize: 15.98GiB deviceMemoryBandwidth: 953.67GiB/s 2021-04-21 10:30:54.256246: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1734] Found device 3 with properties: pciBusID: 0000:63:00.0 name: Device 66a1 ROCm AMD GPU ISA: gfx906 coreClock: 1.7GHz coreCount: 64 deviceMemorySize: 15.98GiB deviceMemoryBandwidth: 953.67GiB/s 2021-04-21 10:30:54.809456: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library librocblas.so 2021-04-21 10:30:55.032434: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libMIOpen.so 2021-04-21 10:30:55.784559: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library librocfft.so 2021-04-21 10:30:55.838353: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library librocrand.so 2021-04-21 10:30:55.839140: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1858] Adding visible gpu devices: 0, 1, 2, 3 2021-04-21 10:30:55.925194: I tensorflow/core/platform/profile_utils/cpu_utils.cc:104] CPU Frequency: 1999885000 Hz 2021-04-21 10:30:55.928441: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x5afeec0 initialized for platform Host (this does not guarantee that XLA will be used). Devices: 2021-04-21 10:30:55.928565: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version 2021-04-21 10:30:55.935554: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x5b6ba20 initialized for platform ROCM (this does not guarantee that XLA will be used). Devices: 2021-04-21 10:30:55.935628: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Device 66a1, AMDGPU ISA version: gfx906 2021-04-21 10:30:55.935667: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (1): Device 66a1, AMDGPU ISA version: gfx906 2021-04-21 10:30:55.935702: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (2): Device 66a1, AMDGPU ISA version: gfx906 2021-04-21 10:30:55.935752: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (3): Device 66a1, AMDGPU ISA version: gfx906 2021-04-21 10:31:00.379226: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1734] Found device 0 with properties: pciBusID: 0000:04:00.0 name: Device 66a1 ROCm AMD GPU ISA: gfx906 coreClock: 1.7GHz coreCount: 64 deviceMemorySize: 15.98GiB deviceMemoryBandwidth: 953.67GiB/s 2021-04-21 10:31:00.379449: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1734] Found device 1 with properties: pciBusID: 0000:26:00.0 name: Device 66a1 ROCm AMD GPU ISA: gfx906 coreClock: 1.7GHz coreCount: 64 deviceMemorySize: 15.98GiB deviceMemoryBandwidth: 953.67GiB/s 2021-04-21 10:31:00.379572: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1734] Found device 2 with properties: pciBusID: 0000:43:00.0 name: Device 66a1 ROCm AMD GPU ISA: gfx906 coreClock: 1.7GHz coreCount: 64 deviceMemorySize: 15.98GiB deviceMemoryBandwidth: 953.67GiB/s 2021-04-21 10:31:00.379700: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1734] Found device 3 with properties: pciBusID: 0000:63:00.0 name: Device 66a1 ROCm AMD GPU ISA: gfx906 coreClock: 1.7GHz coreCount: 64 deviceMemorySize: 15.98GiB deviceMemoryBandwidth: 953.67GiB/s 2021-04-21 10:31:00.379778: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library librocblas.so 2021-04-21 10:31:00.379829: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libMIOpen.so 2021-04-21 10:31:00.379877: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library librocfft.so 2021-04-21 10:31:00.379925: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library librocrand.so 2021-04-21 10:31:00.380484: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1858] Adding visible gpu devices: 0, 1, 2, 3 2021-04-21 10:31:00.380595: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1257] Device interconnect StreamExecutor with strength 1 edge matrix: 2021-04-21 10:31:00.380641: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1263] 0 1 2 3 2021-04-21 10:31:00.380680: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1276] 0: N Y Y Y 2021-04-21 10:31:00.380716: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1276] 1: Y N Y Y 2021-04-21 10:31:00.380752: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1276] 2: Y Y N Y 2021-04-21 10:31:00.380787: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1276] 3: Y Y Y N 2021-04-21 10:31:00.381376: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1402] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 15385 MB memory) -> physical GPU (device: 0, name: Device 66a1, pci bus id: 0000:04:00.0) 2021-04-21 10:31:00.386255: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1402] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:1 with 15385 MB memory) -> physical GPU (device: 1, name: Device 66a1, pci bus id: 0000:26:00.0) 2021-04-21 10:31:00.391170: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1402] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:2 with 15385 MB memory) -> physical GPU (device: 2, name: Device 66a1, pci bus id: 0000:43:00.0) 2021-04-21 10:31:00.395517: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1402] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:3 with 15385 MB memory) -> physical GPU (device: 3, name: Device 66a1, pci bus id: 0000:63:00.0) I0421 10:31:00.405918 47076539613632 mirrored_strategy.py:341] Using MirroredStrategy with devices ('/job:localhost/replica:0/task:0/device:GPU:0', '/job:localhost/replica:0/task:0/device:GPU:1', '/job:localhost/replica:0/task:0/device:GPU:2', '/job:localhost/replica:0/task:0/device:GPU:3') I0421 10:31:00.471962 47076539613632 run_squad_helper.py:236] Training using customized training loop with distribution strategy. W0421 10:31:00.472802 47076539613632 device_compatibility_check.py:111] Mixed precision compatibility check (mixed_float16): WARNING Your GPUs may run slowly with dtype policy mixed_float16 because they do not have compute capability of at least 7.0. Your GPUs: Device 66a1, no compute capability (probably not an Nvidia GPU) (x4) See https://developer.nvidia.com/cuda-gpus for a list of GPUs and their compute capabilities. If you will use compatible GPU(s) not attached to this host, e.g. by running a multi-worker model, you can ignore this warning. This message will only be logged once W0421 10:31:00.473031 47076539613632 deprecation.py:323] From /public/home/xuanbaby/DL-TensorFlow/models_r2.3.0/official/nlp/bert/run_squad_helper.py:295: run_customized_training_loop (from official.nlp.bert.model_training_utils) is deprecated and will be removed in a future version. Instructions for updating: This function is deprecated. Please use Keras compile/fit instead. I0421 10:31:00.473180 47076539613632 model_training_utils.py:228] steps_per_loop not specified. Using steps_per_loop=1 2021-04-21 10:31:01.384660: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 10:31:01.534850: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 10:31:01.542040: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 10:31:01.547573: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 10:31:01.553395: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 10:31:01.558656: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 10:31:01.564427: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 10:31:01.569636: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 10:31:01.575360: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. I0421 10:31:07.485593 47076539613632 optimization.py:89] using Adamw optimizer I0421 10:31:07.494319 47076539613632 model_training_utils.py:273] Checkpoint file /public/home/xuanbaby/DL-TensorFlow/models_r2.3.0/official/nlp/bert/pre_tf2x/bert_model.ckpt found and restoring from initial checkpoint for core model. I0421 10:31:13.980253 47076539613632 model_training_utils.py:276] Loading from checkpoint file completed I0421 10:31:14.110336 47076539613632 cross_device_ops.py:443] Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',). I0421 10:31:14.113797 47076539613632 cross_device_ops.py:443] Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',). I0421 10:31:27.251605 47076539613632 cross_device_ops.py:443] Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',). W0421 10:31:27.837565 47109411751680 optimizer_v2.py:1275] Gradients do not exist for variables ['pooler_transform/kernel:0', 'pooler_transform/bias:0'] when minimizing the loss. W0421 10:31:30.246420 47109409650432 optimizer_v2.py:1275] Gradients do not exist for variables ['pooler_transform/kernel:0', 'pooler_transform/bias:0'] when minimizing the loss. W0421 10:31:30.812205 47109407549184 optimizer_v2.py:1275] Gradients do not exist for variables ['pooler_transform/kernel:0', 'pooler_transform/bias:0'] when minimizing the loss. W0421 10:31:31.555347 47112398051072 optimizer_v2.py:1275] Gradients do not exist for variables ['pooler_transform/kernel:0', 'pooler_transform/bias:0'] when minimizing the loss. I0421 10:31:31.704301 47076539613632 cross_device_ops.py:702] batch_all_reduce: 198 all-reduces with algorithm = nccl, num_packs = 1 W0421 10:31:33.933699 47076539613632 cross_device_ops.py:731] Efficient allreduce is not supported for 1 IndexedSlices I0421 10:31:33.934081 47076539613632 cross_device_ops.py:443] Reduce to /job:localhost/replica:0/task:0/device:GPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:GPU:0', '/job:localhost/replica:0/task:0/device:GPU:1', '/job:localhost/replica:0/task:0/device:GPU:2', '/job:localhost/replica:0/task:0/device:GPU:3'). I0421 10:31:57.373833 47076539613632 cross_device_ops.py:443] Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',). W0421 10:31:57.960069 47109409650432 optimizer_v2.py:1275] Gradients do not exist for variables ['pooler_transform/kernel:0', 'pooler_transform/bias:0'] when minimizing the loss. W0421 10:31:58.537007 47109407549184 optimizer_v2.py:1275] Gradients do not exist for variables ['pooler_transform/kernel:0', 'pooler_transform/bias:0'] when minimizing the loss. W0421 10:31:59.115440 47109411751680 optimizer_v2.py:1275] Gradients do not exist for variables ['pooler_transform/kernel:0', 'pooler_transform/bias:0'] when minimizing the loss. W0421 10:31:59.691951 47112398051072 optimizer_v2.py:1275] Gradients do not exist for variables ['pooler_transform/kernel:0', 'pooler_transform/bias:0'] when minimizing the loss. I0421 10:31:59.841835 47076539613632 cross_device_ops.py:702] batch_all_reduce: 198 all-reduces with algorithm = nccl, num_packs = 1 W0421 10:32:02.086198 47076539613632 cross_device_ops.py:731] Efficient allreduce is not supported for 1 IndexedSlices I0421 10:32:02.086614 47076539613632 cross_device_ops.py:443] Reduce to /job:localhost/replica:0/task:0/device:GPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:GPU:0', '/job:localhost/replica:0/task:0/device:GPU:1', '/job:localhost/replica:0/task:0/device:GPU:2', '/job:localhost/replica:0/task:0/device:GPU:3'). 2021-04-21 10:32:31.860832: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 10:32:33.437892: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library librocblas.so 2021-04-21 10:32:33.628382: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 10:32:33.635352: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. I0421 10:32:48.270071 47076539613632 cross_device_ops.py:443] Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',). I0421 10:32:48.271997 47076539613632 cross_device_ops.py:443] Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',). I0421 10:32:48.298555 47076539613632 model_training_utils.py:505] Train Step: 1/2100 / loss = 5.9443359375 I0421 10:32:48.299917 47076539613632 keras_utils.py:133] TimeHistory: 94.18 seconds, 1.36 examples/second between steps 0 and 1 I0421 10:32:48.321353 47076539613632 cross_device_ops.py:443] Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',). I0421 10:32:48.324232 47076539613632 cross_device_ops.py:443] Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',). I0421 10:32:49.396447 47076539613632 cross_device_ops.py:443] Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',). I0421 10:32:49.398155 47076539613632 cross_device_ops.py:443] Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',). I0421 10:32:49.402400 47076539613632 model_training_utils.py:505] Train Step: 2/2100 / loss = 5.9453125 I0421 10:32:49.402653 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 1 and 2 I0421 10:32:50.484909 47076539613632 model_training_utils.py:505] Train Step: 3/2100 / loss = 5.9326171875 I0421 10:32:50.485297 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 2 and 3 I0421 10:32:51.562718 47076539613632 model_training_utils.py:505] Train Step: 4/2100 / loss = 5.9365234375 I0421 10:32:51.563096 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.71 examples/second between steps 3 and 4 I0421 10:32:52.648627 47076539613632 model_training_utils.py:505] Train Step: 5/2100 / loss = 5.9501953125 I0421 10:32:52.649005 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 4 and 5 I0421 10:32:53.734864 47076539613632 model_training_utils.py:505] Train Step: 6/2100 / loss = 5.951171875 I0421 10:32:53.735242 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 5 and 6 I0421 10:32:54.820068 47076539613632 model_training_utils.py:505] Train Step: 7/2100 / loss = 5.91796875 I0421 10:32:54.820461 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 6 and 7 I0421 10:32:55.906687 47076539613632 model_training_utils.py:505] Train Step: 8/2100 / loss = 5.9443359375 I0421 10:32:55.907068 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 7 and 8 I0421 10:32:56.992115 47076539613632 model_training_utils.py:505] Train Step: 9/2100 / loss = 5.93359375 I0421 10:32:56.992497 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 8 and 9 I0421 10:32:58.077168 47076539613632 model_training_utils.py:505] Train Step: 10/2100 / loss = 5.9287109375 I0421 10:32:58.077557 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 9 and 10 I0421 10:32:59.163030 47076539613632 model_training_utils.py:505] Train Step: 11/2100 / loss = 5.96875 I0421 10:32:59.163427 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 10 and 11 I0421 10:33:00.243196 47076539613632 model_training_utils.py:505] Train Step: 12/2100 / loss = 5.9521484375 I0421 10:33:00.243588 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.44 examples/second between steps 11 and 12 I0421 10:33:01.327911 47076539613632 model_training_utils.py:505] Train Step: 13/2100 / loss = 5.919921875 I0421 10:33:01.328298 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 12 and 13 I0421 10:33:02.410461 47076539613632 model_training_utils.py:505] Train Step: 14/2100 / loss = 5.904296875 I0421 10:33:02.410841 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 13 and 14 I0421 10:33:03.493039 47076539613632 model_training_utils.py:505] Train Step: 15/2100 / loss = 5.9169921875 I0421 10:33:03.493431 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 14 and 15 I0421 10:33:04.580993 47076539613632 model_training_utils.py:505] Train Step: 16/2100 / loss = 5.90625 I0421 10:33:04.581379 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.53 examples/second between steps 15 and 16 I0421 10:33:05.666996 47076539613632 model_training_utils.py:505] Train Step: 17/2100 / loss = 5.9169921875 I0421 10:33:05.667385 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 16 and 17 I0421 10:33:06.754867 47076539613632 model_training_utils.py:505] Train Step: 18/2100 / loss = 5.8935546875 I0421 10:33:06.755258 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.56 examples/second between steps 17 and 18 I0421 10:33:07.838564 47076539613632 model_training_utils.py:505] Train Step: 19/2100 / loss = 5.91015625 I0421 10:33:07.838946 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 18 and 19 I0421 10:33:08.924276 47076539613632 model_training_utils.py:505] Train Step: 20/2100 / loss = 5.890625 I0421 10:33:08.924668 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 19 and 20 I0421 10:33:10.012882 47076539613632 model_training_utils.py:505] Train Step: 21/2100 / loss = 5.900390625 I0421 10:33:10.013268 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.49 examples/second between steps 20 and 21 I0421 10:33:11.104462 47076539613632 model_training_utils.py:505] Train Step: 22/2100 / loss = 5.888671875 I0421 10:33:11.104846 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.15 examples/second between steps 21 and 22 I0421 10:33:12.192351 47076539613632 model_training_utils.py:505] Train Step: 23/2100 / loss = 5.8671875 I0421 10:33:12.192731 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.55 examples/second between steps 22 and 23 I0421 10:33:13.278066 47076539613632 model_training_utils.py:505] Train Step: 24/2100 / loss = 5.880859375 I0421 10:33:13.278454 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 23 and 24 I0421 10:33:14.365594 47076539613632 model_training_utils.py:505] Train Step: 25/2100 / loss = 5.8662109375 I0421 10:33:14.365981 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.59 examples/second between steps 24 and 25 I0421 10:33:15.452601 47076539613632 model_training_utils.py:505] Train Step: 26/2100 / loss = 5.865234375 I0421 10:33:15.452991 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 25 and 26 I0421 10:33:16.539515 47076539613632 model_training_utils.py:505] Train Step: 27/2100 / loss = 5.876953125 I0421 10:33:16.539895 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.62 examples/second between steps 26 and 27 I0421 10:33:17.628303 47076539613632 model_training_utils.py:505] Train Step: 28/2100 / loss = 5.8125 I0421 10:33:17.628740 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.55 examples/second between steps 27 and 28 I0421 10:33:18.727007 47076539613632 model_training_utils.py:505] Train Step: 29/2100 / loss = 5.814453125 I0421 10:33:18.727447 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.40 examples/second between steps 28 and 29 I0421 10:33:19.820886 47076539613632 model_training_utils.py:505] Train Step: 30/2100 / loss = 5.822265625 I0421 10:33:19.821330 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.92 examples/second between steps 29 and 30 I0421 10:33:20.913674 47076539613632 model_training_utils.py:505] Train Step: 31/2100 / loss = 5.7841796875 I0421 10:33:20.914122 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.05 examples/second between steps 30 and 31 I0421 10:33:22.004764 47076539613632 model_training_utils.py:505] Train Step: 32/2100 / loss = 5.7919921875 I0421 10:33:22.005201 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.23 examples/second between steps 31 and 32 I0421 10:33:23.096519 47076539613632 model_training_utils.py:505] Train Step: 33/2100 / loss = 5.7685546875 I0421 10:33:23.096954 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.17 examples/second between steps 32 and 33 I0421 10:33:24.181210 47076539613632 model_training_utils.py:505] Train Step: 34/2100 / loss = 5.7822265625 I0421 10:33:24.181681 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 33 and 34 I0421 10:33:25.272241 47076539613632 model_training_utils.py:505] Train Step: 35/2100 / loss = 5.751953125 I0421 10:33:25.272684 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.26 examples/second between steps 34 and 35 I0421 10:33:26.363500 47076539613632 model_training_utils.py:505] Train Step: 36/2100 / loss = 5.7421875 I0421 10:33:26.363940 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.20 examples/second between steps 35 and 36 I0421 10:33:27.455332 47076539613632 model_training_utils.py:505] Train Step: 37/2100 / loss = 5.724609375 I0421 10:33:27.455775 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.17 examples/second between steps 36 and 37 I0421 10:33:28.539311 47076539613632 model_training_utils.py:505] Train Step: 38/2100 / loss = 5.748046875 I0421 10:33:28.539747 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 37 and 38 I0421 10:33:29.628311 47076539613632 model_training_utils.py:505] Train Step: 39/2100 / loss = 5.724609375 I0421 10:33:29.628753 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.46 examples/second between steps 38 and 39 I0421 10:33:30.714823 47076539613632 model_training_utils.py:505] Train Step: 40/2100 / loss = 5.767578125 I0421 10:33:30.715259 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 39 and 40 I0421 10:33:31.807786 47076539613632 model_training_utils.py:505] Train Step: 41/2100 / loss = 5.748046875 I0421 10:33:31.808224 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.05 examples/second between steps 40 and 41 I0421 10:33:32.893361 47076539613632 model_training_utils.py:505] Train Step: 42/2100 / loss = 5.7373046875 I0421 10:33:32.893798 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 41 and 42 I0421 10:33:33.984423 47076539613632 model_training_utils.py:505] Train Step: 43/2100 / loss = 5.69140625 I0421 10:33:33.984850 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.26 examples/second between steps 42 and 43 I0421 10:33:35.077356 47076539613632 model_training_utils.py:505] Train Step: 44/2100 / loss = 5.6630859375 I0421 10:33:35.077801 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.02 examples/second between steps 43 and 44 I0421 10:33:36.164971 47076539613632 model_training_utils.py:505] Train Step: 45/2100 / loss = 5.63671875 I0421 10:33:36.165435 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.63 examples/second between steps 44 and 45 I0421 10:33:37.251868 47076539613632 model_training_utils.py:505] Train Step: 46/2100 / loss = 5.6240234375 I0421 10:33:37.252318 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 45 and 46 I0421 10:33:38.336815 47076539613632 model_training_utils.py:505] Train Step: 47/2100 / loss = 5.5634765625 I0421 10:33:38.337260 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 46 and 47 I0421 10:33:39.423032 47076539613632 model_training_utils.py:505] Train Step: 48/2100 / loss = 5.56640625 I0421 10:33:39.423485 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 47 and 48 I0421 10:33:40.504391 47076539613632 model_training_utils.py:505] Train Step: 49/2100 / loss = 5.6220703125 I0421 10:33:40.504829 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.31 examples/second between steps 48 and 49 I0421 10:33:41.589461 47076539613632 model_training_utils.py:505] Train Step: 50/2100 / loss = 5.587890625 I0421 10:33:41.589900 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 49 and 50 I0421 10:33:42.675948 47076539613632 model_training_utils.py:505] Train Step: 51/2100 / loss = 5.591796875 I0421 10:33:42.676396 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 50 and 51 I0421 10:33:43.769904 47076539613632 model_training_utils.py:505] Train Step: 52/2100 / loss = 5.501953125 I0421 10:33:43.770345 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.93 examples/second between steps 51 and 52 I0421 10:33:44.866144 47076539613632 model_training_utils.py:505] Train Step: 53/2100 / loss = 5.529296875 I0421 10:33:44.866591 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.69 examples/second between steps 52 and 53 I0421 10:33:45.962362 47076539613632 model_training_utils.py:505] Train Step: 54/2100 / loss = 5.447265625 I0421 10:33:45.962801 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.67 examples/second between steps 53 and 54 I0421 10:33:47.056687 47076539613632 model_training_utils.py:505] Train Step: 55/2100 / loss = 5.359375 I0421 10:33:47.057127 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.90 examples/second between steps 54 and 55 I0421 10:33:48.151849 47076539613632 model_training_utils.py:505] Train Step: 56/2100 / loss = 5.287109375 I0421 10:33:48.152288 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.79 examples/second between steps 55 and 56 I0421 10:33:49.242059 47076539613632 model_training_utils.py:505] Train Step: 57/2100 / loss = 5.318359375 I0421 10:33:49.242503 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.38 examples/second between steps 56 and 57 I0421 10:33:50.339562 47076539613632 model_training_utils.py:505] Train Step: 58/2100 / loss = 5.26171875 I0421 10:33:50.340005 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.53 examples/second between steps 57 and 58 I0421 10:33:51.443152 47076539613632 model_training_utils.py:505] Train Step: 59/2100 / loss = 5.203125 I0421 10:33:51.443610 47076539613632 keras_utils.py:133] TimeHistory: 1.10 seconds, 116.88 examples/second between steps 58 and 59 I0421 10:33:52.547581 47076539613632 model_training_utils.py:505] Train Step: 60/2100 / loss = 5.177734375 I0421 10:33:52.548020 47076539613632 keras_utils.py:133] TimeHistory: 1.10 seconds, 116.78 examples/second between steps 59 and 60 I0421 10:33:53.637635 47076539613632 model_training_utils.py:505] Train Step: 61/2100 / loss = 5.1201171875 I0421 10:33:53.638072 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.34 examples/second between steps 60 and 61 I0421 10:33:54.727105 47076539613632 model_training_utils.py:505] Train Step: 62/2100 / loss = 5.150390625 I0421 10:33:54.727555 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.38 examples/second between steps 61 and 62 I0421 10:33:55.815848 47076539613632 model_training_utils.py:505] Train Step: 63/2100 / loss = 5.130859375 I0421 10:33:55.816307 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.47 examples/second between steps 62 and 63 I0421 10:33:56.908647 47076539613632 model_training_utils.py:505] Train Step: 64/2100 / loss = 5.1796875 I0421 10:33:56.909087 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.03 examples/second between steps 63 and 64 I0421 10:33:58.003484 47076539613632 model_training_utils.py:505] Train Step: 65/2100 / loss = 5.1591796875 I0421 10:33:58.003915 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.82 examples/second between steps 64 and 65 I0421 10:33:59.088458 47076539613632 model_training_utils.py:505] Train Step: 66/2100 / loss = 5.171875 I0421 10:33:59.088898 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 65 and 66 I0421 10:34:00.169981 47076539613632 model_training_utils.py:505] Train Step: 67/2100 / loss = 5.185546875 I0421 10:34:00.170428 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 66 and 67 I0421 10:34:01.251944 47076539613632 model_training_utils.py:505] Train Step: 68/2100 / loss = 5.2021484375 I0421 10:34:01.252393 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 67 and 68 I0421 10:34:02.332823 47076539613632 model_training_utils.py:505] Train Step: 69/2100 / loss = 5.1845703125 I0421 10:34:02.333277 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.34 examples/second between steps 68 and 69 I0421 10:34:03.417798 47076539613632 model_training_utils.py:505] Train Step: 70/2100 / loss = 5.0703125 I0421 10:34:03.418241 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 69 and 70 I0421 10:34:04.506007 47076539613632 model_training_utils.py:505] Train Step: 71/2100 / loss = 4.97265625 I0421 10:34:04.506458 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.57 examples/second between steps 70 and 71 I0421 10:34:05.593928 47076539613632 model_training_utils.py:505] Train Step: 72/2100 / loss = 4.994140625 I0421 10:34:05.594376 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.58 examples/second between steps 71 and 72 I0421 10:34:06.681665 47076539613632 model_training_utils.py:505] Train Step: 73/2100 / loss = 4.84765625 I0421 10:34:06.682108 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.61 examples/second between steps 72 and 73 I0421 10:34:07.769192 47076539613632 model_training_utils.py:505] Train Step: 74/2100 / loss = 4.69921875 I0421 10:34:07.769628 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.63 examples/second between steps 73 and 74 I0421 10:34:08.853453 47076539613632 model_training_utils.py:505] Train Step: 75/2100 / loss = 4.6982421875 I0421 10:34:08.853891 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 74 and 75 I0421 10:34:09.946589 47076539613632 model_training_utils.py:505] Train Step: 76/2100 / loss = 4.580078125 I0421 10:34:09.947052 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.00 examples/second between steps 75 and 76 I0421 10:34:11.047191 47076539613632 model_training_utils.py:505] Train Step: 77/2100 / loss = 4.3583984375 I0421 10:34:11.047638 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.18 examples/second between steps 76 and 77 I0421 10:34:12.143185 47076539613632 model_training_utils.py:505] Train Step: 78/2100 / loss = 4.2734375 I0421 10:34:12.143626 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.68 examples/second between steps 77 and 78 I0421 10:34:13.242194 47076539613632 model_training_utils.py:505] Train Step: 79/2100 / loss = 4.1494140625 I0421 10:34:13.242636 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.36 examples/second between steps 78 and 79 I0421 10:34:14.339607 47076539613632 model_training_utils.py:505] Train Step: 80/2100 / loss = 4.10986328125 I0421 10:34:14.340050 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.58 examples/second between steps 79 and 80 I0421 10:34:15.435370 47076539613632 model_training_utils.py:505] Train Step: 81/2100 / loss = 4.177734375 I0421 10:34:15.435805 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.73 examples/second between steps 80 and 81 I0421 10:34:16.534599 47076539613632 model_training_utils.py:505] Train Step: 82/2100 / loss = 4.24609375 I0421 10:34:16.535030 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.35 examples/second between steps 81 and 82 I0421 10:34:17.630459 47076539613632 model_training_utils.py:505] Train Step: 83/2100 / loss = 4.333984375 I0421 10:34:17.630895 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.72 examples/second between steps 82 and 83 I0421 10:34:18.724392 47076539613632 model_training_utils.py:505] Train Step: 84/2100 / loss = 4.33984375 I0421 10:34:18.724826 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.92 examples/second between steps 83 and 84 I0421 10:34:19.816267 47076539613632 model_training_utils.py:505] Train Step: 85/2100 / loss = 4.14453125 I0421 10:34:19.816704 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.16 examples/second between steps 84 and 85 I0421 10:34:20.912571 47076539613632 model_training_utils.py:505] Train Step: 86/2100 / loss = 4.12890625 I0421 10:34:20.913006 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.67 examples/second between steps 85 and 86 I0421 10:34:22.010862 47076539613632 model_training_utils.py:505] Train Step: 87/2100 / loss = 4.02197265625 I0421 10:34:22.011304 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.47 examples/second between steps 86 and 87 I0421 10:34:23.109011 47076539613632 model_training_utils.py:505] Train Step: 88/2100 / loss = 3.880859375 I0421 10:34:23.109453 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.49 examples/second between steps 87 and 88 I0421 10:34:24.211528 47076539613632 model_training_utils.py:505] Train Step: 89/2100 / loss = 3.6865234375 I0421 10:34:24.211975 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 116.99 examples/second between steps 88 and 89 I0421 10:34:25.310386 47076539613632 model_training_utils.py:505] Train Step: 90/2100 / loss = 3.8046875 I0421 10:34:25.310827 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.39 examples/second between steps 89 and 90 I0421 10:34:26.396572 47076539613632 model_training_utils.py:505] Train Step: 91/2100 / loss = 3.94287109375 I0421 10:34:26.397000 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 90 and 91 I0421 10:34:27.490232 47076539613632 model_training_utils.py:505] Train Step: 92/2100 / loss = 4.060546875 I0421 10:34:27.490672 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.97 examples/second between steps 91 and 92 I0421 10:34:28.581966 47076539613632 model_training_utils.py:505] Train Step: 93/2100 / loss = 3.89501953125 I0421 10:34:28.582418 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.17 examples/second between steps 92 and 93 I0421 10:34:29.672203 47076539613632 model_training_utils.py:505] Train Step: 94/2100 / loss = 3.7763671875 I0421 10:34:29.672649 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.32 examples/second between steps 93 and 94 I0421 10:34:30.760814 47076539613632 model_training_utils.py:505] Train Step: 95/2100 / loss = 3.6962890625 I0421 10:34:30.761252 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.51 examples/second between steps 94 and 95 I0421 10:34:31.854213 47076539613632 model_training_utils.py:505] Train Step: 96/2100 / loss = 3.724609375 I0421 10:34:31.854670 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.97 examples/second between steps 95 and 96 I0421 10:34:32.938133 47076539613632 model_training_utils.py:505] Train Step: 97/2100 / loss = 3.7607421875 I0421 10:34:32.938577 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 96 and 97 I0421 10:34:34.027951 47076539613632 model_training_utils.py:505] Train Step: 98/2100 / loss = 3.64453125 I0421 10:34:34.028393 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.38 examples/second between steps 97 and 98 I0421 10:34:35.117880 47076539613632 model_training_utils.py:505] Train Step: 99/2100 / loss = 3.72607421875 I0421 10:34:35.118323 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.36 examples/second between steps 98 and 99 I0421 10:34:36.212108 47076539613632 model_training_utils.py:505] Train Step: 100/2100 / loss = 3.4716796875 I0421 10:34:36.212553 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.91 examples/second between steps 99 and 100 I0421 10:34:37.305122 47076539613632 model_training_utils.py:505] Train Step: 101/2100 / loss = 3.4677734375 I0421 10:34:37.305563 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.07 examples/second between steps 100 and 101 I0421 10:34:38.400527 47076539613632 model_training_utils.py:505] Train Step: 102/2100 / loss = 3.6123046875 I0421 10:34:38.400963 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.77 examples/second between steps 101 and 102 I0421 10:34:39.495639 47076539613632 model_training_utils.py:505] Train Step: 103/2100 / loss = 3.34814453125 I0421 10:34:39.496074 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.82 examples/second between steps 102 and 103 I0421 10:34:40.591315 47076539613632 model_training_utils.py:505] Train Step: 104/2100 / loss = 3.3125 I0421 10:34:40.591757 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.73 examples/second between steps 103 and 104 I0421 10:34:41.690109 47076539613632 model_training_utils.py:505] Train Step: 105/2100 / loss = 3.4462890625 I0421 10:34:41.690551 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.37 examples/second between steps 104 and 105 I0421 10:34:42.787414 47076539613632 model_training_utils.py:505] Train Step: 106/2100 / loss = 3.361328125 I0421 10:34:42.787856 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.55 examples/second between steps 105 and 106 I0421 10:34:43.880080 47076539613632 model_training_utils.py:505] Train Step: 107/2100 / loss = 3.40771484375 I0421 10:34:43.880532 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.04 examples/second between steps 106 and 107 I0421 10:34:44.970167 47076539613632 model_training_utils.py:505] Train Step: 108/2100 / loss = 3.5498046875 I0421 10:34:44.970622 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.32 examples/second between steps 107 and 108 I0421 10:34:46.058837 47076539613632 model_training_utils.py:505] Train Step: 109/2100 / loss = 3.55224609375 I0421 10:34:46.059267 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.46 examples/second between steps 108 and 109 I0421 10:34:47.146686 47076539613632 model_training_utils.py:505] Train Step: 110/2100 / loss = 3.6591796875 I0421 10:34:47.147127 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.56 examples/second between steps 109 and 110 I0421 10:34:48.237298 47076539613632 model_training_utils.py:505] Train Step: 111/2100 / loss = 3.4384765625 I0421 10:34:48.237743 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.28 examples/second between steps 110 and 111 I0421 10:34:49.331591 47076539613632 model_training_utils.py:505] Train Step: 112/2100 / loss = 3.52392578125 I0421 10:34:49.332030 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.86 examples/second between steps 111 and 112 I0421 10:34:50.422046 47076539613632 model_training_utils.py:505] Train Step: 113/2100 / loss = 3.63720703125 I0421 10:34:50.422496 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.27 examples/second between steps 112 and 113 I0421 10:34:51.519348 47076539613632 model_training_utils.py:505] Train Step: 114/2100 / loss = 3.17041015625 I0421 10:34:51.519803 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.54 examples/second between steps 113 and 114 I0421 10:34:52.624500 47076539613632 model_training_utils.py:505] Train Step: 115/2100 / loss = 3.16259765625 I0421 10:34:52.624935 47076539613632 keras_utils.py:133] TimeHistory: 1.10 seconds, 116.73 examples/second between steps 114 and 115 I0421 10:34:53.721354 47076539613632 model_training_utils.py:505] Train Step: 116/2100 / loss = 3.3212890625 I0421 10:34:53.721795 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.60 examples/second between steps 115 and 116 I0421 10:34:54.820021 47076539613632 model_training_utils.py:505] Train Step: 117/2100 / loss = 3.27587890625 I0421 10:34:54.820466 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.43 examples/second between steps 116 and 117 I0421 10:34:55.917998 47076539613632 model_training_utils.py:505] Train Step: 118/2100 / loss = 3.27587890625 I0421 10:34:55.918444 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.47 examples/second between steps 117 and 118 I0421 10:34:57.004701 47076539613632 model_training_utils.py:505] Train Step: 119/2100 / loss = 3.50048828125 I0421 10:34:57.005135 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.70 examples/second between steps 118 and 119 I0421 10:34:58.094247 47076539613632 model_training_utils.py:505] Train Step: 120/2100 / loss = 3.517578125 I0421 10:34:58.094697 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.39 examples/second between steps 119 and 120 I0421 10:34:59.192052 47076539613632 model_training_utils.py:505] Train Step: 121/2100 / loss = 3.173828125 I0421 10:34:59.192493 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.48 examples/second between steps 120 and 121 I0421 10:35:00.288325 47076539613632 model_training_utils.py:505] Train Step: 122/2100 / loss = 3.3505859375 I0421 10:35:00.288765 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.66 examples/second between steps 121 and 122 I0421 10:35:01.386429 47076539613632 model_training_utils.py:505] Train Step: 123/2100 / loss = 3.208984375 I0421 10:35:01.386867 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.46 examples/second between steps 122 and 123 I0421 10:35:02.486585 47076539613632 model_training_utils.py:505] Train Step: 124/2100 / loss = 3.197265625 I0421 10:35:02.487018 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.26 examples/second between steps 123 and 124 I0421 10:35:03.593462 47076539613632 model_training_utils.py:505] Train Step: 125/2100 / loss = 3.03369140625 I0421 10:35:03.593901 47076539613632 keras_utils.py:133] TimeHistory: 1.10 seconds, 116.51 examples/second between steps 124 and 125 I0421 10:35:04.701143 47076539613632 model_training_utils.py:505] Train Step: 126/2100 / loss = 2.90771484375 I0421 10:35:04.701593 47076539613632 keras_utils.py:133] TimeHistory: 1.10 seconds, 116.43 examples/second between steps 125 and 126 I0421 10:35:05.801792 47076539613632 model_training_utils.py:505] Train Step: 127/2100 / loss = 2.9501953125 I0421 10:35:05.802234 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.19 examples/second between steps 126 and 127 I0421 10:35:06.904639 47076539613632 model_training_utils.py:505] Train Step: 128/2100 / loss = 3.0849609375 I0421 10:35:06.905081 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 116.96 examples/second between steps 127 and 128 I0421 10:35:07.994532 47076539613632 model_training_utils.py:505] Train Step: 129/2100 / loss = 2.919921875 I0421 10:35:07.994969 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.37 examples/second between steps 128 and 129 I0421 10:35:09.090296 47076539613632 model_training_utils.py:505] Train Step: 130/2100 / loss = 2.97021484375 I0421 10:35:09.090724 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.72 examples/second between steps 129 and 130 I0421 10:35:10.189189 47076539613632 model_training_utils.py:505] Train Step: 131/2100 / loss = 2.72900390625 I0421 10:35:10.189634 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.40 examples/second between steps 130 and 131 I0421 10:35:11.283447 47076539613632 model_training_utils.py:505] Train Step: 132/2100 / loss = 2.85693359375 I0421 10:35:11.283894 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.88 examples/second between steps 131 and 132 I0421 10:35:12.377050 47076539613632 model_training_utils.py:505] Train Step: 133/2100 / loss = 2.880859375 I0421 10:35:12.377496 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.94 examples/second between steps 132 and 133 I0421 10:35:13.469250 47076539613632 model_training_utils.py:505] Train Step: 134/2100 / loss = 3.15234375 I0421 10:35:13.469692 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.12 examples/second between steps 133 and 134 I0421 10:35:14.556288 47076539613632 model_training_utils.py:505] Train Step: 135/2100 / loss = 3.20263671875 I0421 10:35:14.556720 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 134 and 135 I0421 10:35:15.646571 47076539613632 model_training_utils.py:505] Train Step: 136/2100 / loss = 3.14892578125 I0421 10:35:15.647006 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.32 examples/second between steps 135 and 136 I0421 10:35:16.744681 47076539613632 model_training_utils.py:505] Train Step: 137/2100 / loss = 2.73291015625 I0421 10:35:16.745126 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.45 examples/second between steps 136 and 137 I0421 10:35:17.839482 47076539613632 model_training_utils.py:505] Train Step: 138/2100 / loss = 2.64404296875 I0421 10:35:17.839912 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.84 examples/second between steps 137 and 138 I0421 10:35:18.936151 47076539613632 model_training_utils.py:505] Train Step: 139/2100 / loss = 2.7490234375 I0421 10:35:18.936597 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.59 examples/second between steps 138 and 139 I0421 10:35:20.035408 47076539613632 model_training_utils.py:505] Train Step: 140/2100 / loss = 2.8759765625 I0421 10:35:20.035847 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.33 examples/second between steps 139 and 140 I0421 10:35:21.133386 47076539613632 model_training_utils.py:505] Train Step: 141/2100 / loss = 2.7998046875 I0421 10:35:21.133822 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.48 examples/second between steps 140 and 141 I0421 10:35:22.235810 47076539613632 model_training_utils.py:505] Train Step: 142/2100 / loss = 2.81640625 I0421 10:35:22.236242 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 116.98 examples/second between steps 141 and 142 I0421 10:35:23.338584 47076539613632 model_training_utils.py:505] Train Step: 143/2100 / loss = 2.73681640625 I0421 10:35:23.339012 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 116.97 examples/second between steps 142 and 143 I0421 10:35:24.440233 47076539613632 model_training_utils.py:505] Train Step: 144/2100 / loss = 2.51513671875 I0421 10:35:24.440684 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.09 examples/second between steps 143 and 144 I0421 10:35:25.534200 47076539613632 model_training_utils.py:505] Train Step: 145/2100 / loss = 2.6640625 I0421 10:35:25.534649 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.94 examples/second between steps 144 and 145 I0421 10:35:26.621137 47076539613632 model_training_utils.py:505] Train Step: 146/2100 / loss = 2.505859375 I0421 10:35:26.621587 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.67 examples/second between steps 145 and 146 I0421 10:35:27.712854 47076539613632 model_training_utils.py:505] Train Step: 147/2100 / loss = 2.638671875 I0421 10:35:27.713296 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.15 examples/second between steps 146 and 147 I0421 10:35:28.808706 47076539613632 model_training_utils.py:505] Train Step: 148/2100 / loss = 2.4208984375 I0421 10:35:28.809145 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.74 examples/second between steps 147 and 148 I0421 10:35:29.898995 47076539613632 model_training_utils.py:505] Train Step: 149/2100 / loss = 2.4814453125 I0421 10:35:29.899448 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.32 examples/second between steps 148 and 149 I0421 10:35:30.984569 47076539613632 model_training_utils.py:505] Train Step: 150/2100 / loss = 2.48876953125 I0421 10:35:30.985006 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 149 and 150 I0421 10:35:32.078787 47076539613632 model_training_utils.py:505] Train Step: 151/2100 / loss = 2.05126953125 I0421 10:35:32.079227 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.88 examples/second between steps 150 and 151 I0421 10:35:33.174210 47076539613632 model_training_utils.py:505] Train Step: 152/2100 / loss = 2.05859375 I0421 10:35:33.174657 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.77 examples/second between steps 151 and 152 I0421 10:35:34.272821 47076539613632 model_training_utils.py:505] Train Step: 153/2100 / loss = 2.203125 I0421 10:35:34.273270 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.40 examples/second between steps 152 and 153 I0421 10:35:35.373723 47076539613632 model_training_utils.py:505] Train Step: 154/2100 / loss = 2.36669921875 I0421 10:35:35.374161 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.18 examples/second between steps 153 and 154 I0421 10:35:36.472821 47076539613632 model_training_utils.py:505] Train Step: 155/2100 / loss = 2.64501953125 I0421 10:35:36.473256 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.39 examples/second between steps 154 and 155 I0421 10:35:37.567892 47076539613632 model_training_utils.py:505] Train Step: 156/2100 / loss = 2.6416015625 I0421 10:35:37.568335 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.81 examples/second between steps 155 and 156 I0421 10:35:38.656911 47076539613632 model_training_utils.py:505] Train Step: 157/2100 / loss = 2.3388671875 I0421 10:35:38.657365 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.44 examples/second between steps 156 and 157 I0421 10:35:39.746025 47076539613632 model_training_utils.py:505] Train Step: 158/2100 / loss = 2.3974609375 I0421 10:35:39.746474 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.44 examples/second between steps 157 and 158 I0421 10:35:40.843460 47076539613632 model_training_utils.py:505] Train Step: 159/2100 / loss = 2.3994140625 I0421 10:35:40.843899 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.54 examples/second between steps 158 and 159 I0421 10:35:41.942160 47076539613632 model_training_utils.py:505] Train Step: 160/2100 / loss = 2.3125 I0421 10:35:41.942608 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.42 examples/second between steps 159 and 160 I0421 10:35:43.037592 47076539613632 model_training_utils.py:505] Train Step: 161/2100 / loss = 2.474609375 I0421 10:35:43.038023 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.76 examples/second between steps 160 and 161 I0421 10:35:44.129074 47076539613632 model_training_utils.py:505] Train Step: 162/2100 / loss = 3.03076171875 I0421 10:35:44.129523 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.17 examples/second between steps 161 and 162 I0421 10:35:45.218172 47076539613632 model_training_utils.py:505] Train Step: 163/2100 / loss = 2.873046875 I0421 10:35:45.218612 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.44 examples/second between steps 162 and 163 I0421 10:35:46.303110 47076539613632 model_training_utils.py:505] Train Step: 164/2100 / loss = 2.30419921875 I0421 10:35:46.303547 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 163 and 164 I0421 10:35:47.392338 47076539613632 model_training_utils.py:505] Train Step: 165/2100 / loss = 2.318359375 I0421 10:35:47.392777 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.42 examples/second between steps 164 and 165 I0421 10:35:48.482771 47076539613632 model_training_utils.py:505] Train Step: 166/2100 / loss = 1.911865234375 I0421 10:35:48.483205 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.27 examples/second between steps 165 and 166 I0421 10:35:49.571064 47076539613632 model_training_utils.py:505] Train Step: 167/2100 / loss = 1.981201171875 I0421 10:35:49.571510 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.50 examples/second between steps 166 and 167 I0421 10:35:50.662383 47076539613632 model_training_utils.py:505] Train Step: 168/2100 / loss = 2.23876953125 I0421 10:35:50.662821 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.18 examples/second between steps 167 and 168 I0421 10:35:51.757626 47076539613632 model_training_utils.py:505] Train Step: 169/2100 / loss = 1.921875 I0421 10:35:51.758062 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.74 examples/second between steps 168 and 169 I0421 10:35:52.860630 47076539613632 model_training_utils.py:505] Train Step: 170/2100 / loss = 1.87451171875 I0421 10:35:52.861064 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 116.93 examples/second between steps 169 and 170 I0421 10:35:53.971435 47076539613632 model_training_utils.py:505] Train Step: 171/2100 / loss = 1.97216796875 I0421 10:35:53.971843 47076539613632 keras_utils.py:133] TimeHistory: 1.10 seconds, 116.10 examples/second between steps 170 and 171 I0421 10:35:55.076320 47076539613632 model_training_utils.py:505] Train Step: 172/2100 / loss = 2.0751953125 I0421 10:35:55.076759 47076539613632 keras_utils.py:133] TimeHistory: 1.10 seconds, 116.71 examples/second between steps 171 and 172 I0421 10:35:56.171914 47076539613632 model_training_utils.py:505] Train Step: 173/2100 / loss = 2.2607421875 I0421 10:35:56.172364 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.72 examples/second between steps 172 and 173 I0421 10:35:57.265091 47076539613632 model_training_utils.py:505] Train Step: 174/2100 / loss = 2.108642578125 I0421 10:35:57.265536 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.02 examples/second between steps 173 and 174 I0421 10:35:58.357677 47076539613632 model_training_utils.py:505] Train Step: 175/2100 / loss = 1.859375 I0421 10:35:58.358111 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.04 examples/second between steps 174 and 175 I0421 10:35:59.447927 47076539613632 model_training_utils.py:505] Train Step: 176/2100 / loss = 1.919189453125 I0421 10:35:59.448372 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.30 examples/second between steps 175 and 176 I0421 10:36:00.539553 47076539613632 model_training_utils.py:505] Train Step: 177/2100 / loss = 1.9541015625 I0421 10:36:00.539986 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.17 examples/second between steps 176 and 177 I0421 10:36:01.638194 47076539613632 model_training_utils.py:505] Train Step: 178/2100 / loss = 1.75390625 I0421 10:36:01.638634 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.38 examples/second between steps 177 and 178 I0421 10:36:02.739884 47076539613632 model_training_utils.py:505] Train Step: 179/2100 / loss = 2.08203125 I0421 10:36:02.740333 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.06 examples/second between steps 178 and 179 I0421 10:36:03.841526 47076539613632 model_training_utils.py:505] Train Step: 180/2100 / loss = 2.353515625 I0421 10:36:03.841967 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.08 examples/second between steps 179 and 180 I0421 10:36:04.941748 47076539613632 model_training_utils.py:505] Train Step: 181/2100 / loss = 2.74755859375 I0421 10:36:04.942192 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.22 examples/second between steps 180 and 181 I0421 10:36:06.044015 47076539613632 model_training_utils.py:505] Train Step: 182/2100 / loss = 2.58251953125 I0421 10:36:06.044461 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.01 examples/second between steps 181 and 182 I0421 10:36:07.136699 47076539613632 model_training_utils.py:505] Train Step: 183/2100 / loss = 2.5166015625 I0421 10:36:07.137087 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.08 examples/second between steps 182 and 183 I0421 10:36:08.227677 47076539613632 model_training_utils.py:505] Train Step: 184/2100 / loss = 2.206298828125 I0421 10:36:08.228062 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.21 examples/second between steps 183 and 184 I0421 10:36:09.317612 47076539613632 model_training_utils.py:505] Train Step: 185/2100 / loss = 2.0595703125 I0421 10:36:09.318001 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.38 examples/second between steps 184 and 185 I0421 10:36:10.399769 47076539613632 model_training_utils.py:505] Train Step: 186/2100 / loss = 2.02099609375 I0421 10:36:10.400155 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 185 and 186 I0421 10:36:11.481112 47076539613632 model_training_utils.py:505] Train Step: 187/2100 / loss = 1.963623046875 I0421 10:36:11.481510 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.47 examples/second between steps 186 and 187 I0421 10:36:12.561995 47076539613632 model_training_utils.py:505] Train Step: 188/2100 / loss = 1.969482421875 I0421 10:36:12.562394 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.32 examples/second between steps 187 and 188 I0421 10:36:13.648635 47076539613632 model_training_utils.py:505] Train Step: 189/2100 / loss = 2.1064453125 I0421 10:36:13.649017 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 188 and 189 I0421 10:36:14.748042 47076539613632 model_training_utils.py:505] Train Step: 190/2100 / loss = 1.982421875 I0421 10:36:14.748435 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.32 examples/second between steps 189 and 190 I0421 10:36:15.846697 47076539613632 model_training_utils.py:505] Train Step: 191/2100 / loss = 2.490234375 I0421 10:36:15.847087 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.42 examples/second between steps 190 and 191 I0421 10:36:16.945017 47076539613632 model_training_utils.py:505] Train Step: 192/2100 / loss = 2.7373046875 I0421 10:36:16.945409 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.43 examples/second between steps 191 and 192 I0421 10:36:18.035637 47076539613632 model_training_utils.py:505] Train Step: 193/2100 / loss = 3.0576171875 I0421 10:36:18.036022 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.26 examples/second between steps 192 and 193 I0421 10:36:19.131561 47076539613632 model_training_utils.py:505] Train Step: 194/2100 / loss = 2.26318359375 I0421 10:36:19.131948 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.69 examples/second between steps 193 and 194 I0421 10:36:20.230017 47076539613632 model_training_utils.py:505] Train Step: 195/2100 / loss = 1.943115234375 I0421 10:36:20.230407 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.40 examples/second between steps 194 and 195 I0421 10:36:21.328803 47076539613632 model_training_utils.py:505] Train Step: 196/2100 / loss = 1.876220703125 I0421 10:36:21.329184 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.38 examples/second between steps 195 and 196 I0421 10:36:22.421467 47076539613632 model_training_utils.py:505] Train Step: 197/2100 / loss = 2.02978515625 I0421 10:36:22.421845 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.05 examples/second between steps 196 and 197 I0421 10:36:23.515832 47076539613632 model_training_utils.py:505] Train Step: 198/2100 / loss = 2.3974609375 I0421 10:36:23.516227 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.86 examples/second between steps 197 and 198 I0421 10:36:24.613873 47076539613632 model_training_utils.py:505] Train Step: 199/2100 / loss = 2.5244140625 I0421 10:36:24.614255 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.47 examples/second between steps 198 and 199 I0421 10:36:25.708716 47076539613632 model_training_utils.py:505] Train Step: 200/2100 / loss = 1.859130859375 I0421 10:36:25.709094 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.83 examples/second between steps 199 and 200 I0421 10:36:26.797033 47076539613632 model_training_utils.py:505] Train Step: 201/2100 / loss = 1.78759765625 I0421 10:36:26.797431 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.51 examples/second between steps 200 and 201 I0421 10:36:27.890554 47076539613632 model_training_utils.py:505] Train Step: 202/2100 / loss = 2.07958984375 I0421 10:36:27.890937 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.93 examples/second between steps 201 and 202 I0421 10:36:28.984051 47076539613632 model_training_utils.py:505] Train Step: 203/2100 / loss = 2.04736328125 I0421 10:36:28.984441 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.94 examples/second between steps 202 and 203 I0421 10:36:30.081305 47076539613632 model_training_utils.py:505] Train Step: 204/2100 / loss = 2.20654296875 I0421 10:36:30.081685 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.54 examples/second between steps 203 and 204 I0421 10:36:31.172506 47076539613632 model_training_utils.py:505] Train Step: 205/2100 / loss = 2.078369140625 I0421 10:36:31.172886 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.21 examples/second between steps 204 and 205 I0421 10:36:32.263342 47076539613632 model_training_utils.py:505] Train Step: 206/2100 / loss = 1.963623046875 I0421 10:36:32.263725 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.24 examples/second between steps 205 and 206 I0421 10:36:33.348359 47076539613632 model_training_utils.py:505] Train Step: 207/2100 / loss = 1.90673828125 I0421 10:36:33.348739 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 206 and 207 I0421 10:36:34.432911 47076539613632 model_training_utils.py:505] Train Step: 208/2100 / loss = 1.884765625 I0421 10:36:34.433306 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 207 and 208 I0421 10:36:35.520555 47076539613632 model_training_utils.py:505] Train Step: 209/2100 / loss = 1.528076171875 I0421 10:36:35.520941 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.62 examples/second between steps 208 and 209 I0421 10:36:36.608789 47076539613632 model_training_utils.py:505] Train Step: 210/2100 / loss = 1.675537109375 I0421 10:36:36.609187 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.55 examples/second between steps 209 and 210 I0421 10:36:37.698480 47076539613632 model_training_utils.py:505] Train Step: 211/2100 / loss = 1.9228515625 I0421 10:36:37.698879 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.37 examples/second between steps 210 and 211 I0421 10:36:38.783445 47076539613632 model_training_utils.py:505] Train Step: 212/2100 / loss = 2.302734375 I0421 10:36:38.783826 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 211 and 212 I0421 10:36:39.875589 47076539613632 model_training_utils.py:505] Train Step: 213/2100 / loss = 2.113037109375 I0421 10:36:39.875974 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.10 examples/second between steps 212 and 213 I0421 10:36:40.966915 47076539613632 model_training_utils.py:505] Train Step: 214/2100 / loss = 2.05712890625 I0421 10:36:40.967299 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.19 examples/second between steps 213 and 214 I0421 10:36:42.054537 47076539613632 model_training_utils.py:505] Train Step: 215/2100 / loss = 2.138671875 I0421 10:36:42.054910 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.59 examples/second between steps 214 and 215 I0421 10:36:43.147522 47076539613632 model_training_utils.py:505] Train Step: 216/2100 / loss = 2.02392578125 I0421 10:36:43.147897 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.03 examples/second between steps 215 and 216 I0421 10:36:44.236444 47076539613632 model_training_utils.py:505] Train Step: 217/2100 / loss = 1.916259765625 I0421 10:36:44.236830 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.46 examples/second between steps 216 and 217 I0421 10:36:45.320531 47076539613632 model_training_utils.py:505] Train Step: 218/2100 / loss = 1.693359375 I0421 10:36:45.320907 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 217 and 218 I0421 10:36:46.407844 47076539613632 model_training_utils.py:505] Train Step: 219/2100 / loss = 1.822265625 I0421 10:36:46.408226 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.63 examples/second between steps 218 and 219 I0421 10:36:47.494579 47076539613632 model_training_utils.py:505] Train Step: 220/2100 / loss = 2.04150390625 I0421 10:36:47.494957 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 219 and 220 I0421 10:36:48.580131 47076539613632 model_training_utils.py:505] Train Step: 221/2100 / loss = 1.9169921875 I0421 10:36:48.580522 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 220 and 221 I0421 10:36:49.667216 47076539613632 model_training_utils.py:505] Train Step: 222/2100 / loss = 1.70849609375 I0421 10:36:49.667606 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 221 and 222 I0421 10:36:50.753705 47076539613632 model_training_utils.py:505] Train Step: 223/2100 / loss = 2.052001953125 I0421 10:36:50.754083 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 222 and 223 I0421 10:36:51.843742 47076539613632 model_training_utils.py:505] Train Step: 224/2100 / loss = 1.72412109375 I0421 10:36:51.844120 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.31 examples/second between steps 223 and 224 I0421 10:36:52.930403 47076539613632 model_training_utils.py:505] Train Step: 225/2100 / loss = 1.67333984375 I0421 10:36:52.930780 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 224 and 225 I0421 10:36:54.013795 47076539613632 model_training_utils.py:505] Train Step: 226/2100 / loss = 1.578857421875 I0421 10:36:54.014178 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 225 and 226 I0421 10:36:55.094019 47076539613632 model_training_utils.py:505] Train Step: 227/2100 / loss = 1.9794921875 I0421 10:36:55.094408 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.40 examples/second between steps 226 and 227 I0421 10:36:56.181170 47076539613632 model_training_utils.py:505] Train Step: 228/2100 / loss = 1.74462890625 I0421 10:36:56.181609 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 227 and 228 I0421 10:36:57.272562 47076539613632 model_training_utils.py:505] Train Step: 229/2100 / loss = 1.9248046875 I0421 10:36:57.273015 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.19 examples/second between steps 228 and 229 I0421 10:36:58.364504 47076539613632 model_training_utils.py:505] Train Step: 230/2100 / loss = 2.0712890625 I0421 10:36:58.364936 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.18 examples/second between steps 229 and 230 I0421 10:36:59.452755 47076539613632 model_training_utils.py:505] Train Step: 231/2100 / loss = 1.64599609375 I0421 10:36:59.453196 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.52 examples/second between steps 230 and 231 I0421 10:37:00.538913 47076539613632 model_training_utils.py:505] Train Step: 232/2100 / loss = 1.5927734375 I0421 10:37:00.539364 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 231 and 232 I0421 10:37:01.629975 47076539613632 model_training_utils.py:505] Train Step: 233/2100 / loss = 2.10498046875 I0421 10:37:01.630430 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.22 examples/second between steps 232 and 233 I0421 10:37:02.720572 47076539613632 model_training_utils.py:505] Train Step: 234/2100 / loss = 2.256591796875 I0421 10:37:02.721011 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.26 examples/second between steps 233 and 234 I0421 10:37:03.807444 47076539613632 model_training_utils.py:505] Train Step: 235/2100 / loss = 2.27099609375 I0421 10:37:03.807877 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.69 examples/second between steps 234 and 235 I0421 10:37:04.892779 47076539613632 model_training_utils.py:505] Train Step: 236/2100 / loss = 2.05224609375 I0421 10:37:04.893216 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 235 and 236 I0421 10:37:05.972440 47076539613632 model_training_utils.py:505] Train Step: 237/2100 / loss = 2.333251953125 I0421 10:37:05.972879 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.47 examples/second between steps 236 and 237 I0421 10:37:07.055802 47076539613632 model_training_utils.py:505] Train Step: 238/2100 / loss = 1.96630859375 I0421 10:37:07.056239 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 237 and 238 I0421 10:37:08.144304 47076539613632 model_training_utils.py:505] Train Step: 239/2100 / loss = 1.80517578125 I0421 10:37:08.144689 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 238 and 239 I0421 10:37:09.233993 47076539613632 model_training_utils.py:505] Train Step: 240/2100 / loss = 1.977294921875 I0421 10:37:09.234385 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.37 examples/second between steps 239 and 240 I0421 10:37:10.324296 47076539613632 model_training_utils.py:505] Train Step: 241/2100 / loss = 2.26953125 I0421 10:37:10.324686 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.29 examples/second between steps 240 and 241 I0421 10:37:11.411832 47076539613632 model_training_utils.py:505] Train Step: 242/2100 / loss = 2.10498046875 I0421 10:37:11.412266 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 241 and 242 I0421 10:37:12.497775 47076539613632 model_training_utils.py:505] Train Step: 243/2100 / loss = 2.392578125 I0421 10:37:12.498204 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 242 and 243 I0421 10:37:13.586014 47076539613632 model_training_utils.py:505] Train Step: 244/2100 / loss = 2.3408203125 I0421 10:37:13.586451 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.56 examples/second between steps 243 and 244 I0421 10:37:14.671746 47076539613632 model_training_utils.py:505] Train Step: 245/2100 / loss = 2.14453125 I0421 10:37:14.672190 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 244 and 245 I0421 10:37:15.757756 47076539613632 model_training_utils.py:505] Train Step: 246/2100 / loss = 2.69482421875 I0421 10:37:15.758199 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 245 and 246 I0421 10:37:16.846297 47076539613632 model_training_utils.py:505] Train Step: 247/2100 / loss = 2.28076171875 I0421 10:37:16.846732 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.49 examples/second between steps 246 and 247 I0421 10:37:17.933180 47076539613632 model_training_utils.py:505] Train Step: 248/2100 / loss = 1.96826171875 I0421 10:37:17.933619 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 247 and 248 I0421 10:37:19.019651 47076539613632 model_training_utils.py:505] Train Step: 249/2100 / loss = 2.0517578125 I0421 10:37:19.020088 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 248 and 249 I0421 10:37:20.105462 47076539613632 model_training_utils.py:505] Train Step: 250/2100 / loss = 1.95849609375 I0421 10:37:20.105892 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 249 and 250 I0421 10:37:21.195276 47076539613632 model_training_utils.py:505] Train Step: 251/2100 / loss = 1.484619140625 I0421 10:37:21.195712 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.35 examples/second between steps 250 and 251 I0421 10:37:22.281736 47076539613632 model_training_utils.py:505] Train Step: 252/2100 / loss = 1.4638671875 I0421 10:37:22.282172 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 251 and 252 I0421 10:37:23.361393 47076539613632 model_training_utils.py:505] Train Step: 253/2100 / loss = 1.204833984375 I0421 10:37:23.361828 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.46 examples/second between steps 252 and 253 I0421 10:37:24.446151 47076539613632 model_training_utils.py:505] Train Step: 254/2100 / loss = 1.101318359375 I0421 10:37:24.446596 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 253 and 254 I0421 10:37:25.531286 47076539613632 model_training_utils.py:505] Train Step: 255/2100 / loss = 1.52685546875 I0421 10:37:25.531720 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 254 and 255 I0421 10:37:26.611370 47076539613632 model_training_utils.py:505] Train Step: 256/2100 / loss = 1.5205078125 I0421 10:37:26.611806 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.41 examples/second between steps 255 and 256 I0421 10:37:27.699063 47076539613632 model_training_utils.py:505] Train Step: 257/2100 / loss = 1.759521484375 I0421 10:37:27.699530 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.58 examples/second between steps 256 and 257 I0421 10:37:28.783625 47076539613632 model_training_utils.py:505] Train Step: 258/2100 / loss = 1.6201171875 I0421 10:37:28.784062 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 257 and 258 I0421 10:37:29.872351 47076539613632 model_training_utils.py:505] Train Step: 259/2100 / loss = 1.42236328125 I0421 10:37:29.872783 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.47 examples/second between steps 258 and 259 I0421 10:37:30.960450 47076539613632 model_training_utils.py:505] Train Step: 260/2100 / loss = 1.73095703125 I0421 10:37:30.960888 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.53 examples/second between steps 259 and 260 I0421 10:37:32.042728 47076539613632 model_training_utils.py:505] Train Step: 261/2100 / loss = 1.8154296875 I0421 10:37:32.043156 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 260 and 261 I0421 10:37:33.124570 47076539613632 model_training_utils.py:505] Train Step: 262/2100 / loss = 1.322021484375 I0421 10:37:33.125001 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 261 and 262 I0421 10:37:34.209725 47076539613632 model_training_utils.py:505] Train Step: 263/2100 / loss = 1.29345703125 I0421 10:37:34.210155 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 262 and 263 I0421 10:37:35.294511 47076539613632 model_training_utils.py:505] Train Step: 264/2100 / loss = 1.447998046875 I0421 10:37:35.294948 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 263 and 264 I0421 10:37:36.379417 47076539613632 model_training_utils.py:505] Train Step: 265/2100 / loss = 1.773681640625 I0421 10:37:36.379853 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 264 and 265 I0421 10:37:37.465814 47076539613632 model_training_utils.py:505] Train Step: 266/2100 / loss = 1.57421875 I0421 10:37:37.466248 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 265 and 266 I0421 10:37:38.555461 47076539613632 model_training_utils.py:505] Train Step: 267/2100 / loss = 1.718505859375 I0421 10:37:38.555895 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.39 examples/second between steps 266 and 267 I0421 10:37:39.643021 47076539613632 model_training_utils.py:505] Train Step: 268/2100 / loss = 1.618896484375 I0421 10:37:39.643476 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.61 examples/second between steps 267 and 268 I0421 10:37:40.727110 47076539613632 model_training_utils.py:505] Train Step: 269/2100 / loss = 1.542724609375 I0421 10:37:40.727553 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 268 and 269 I0421 10:37:41.814102 47076539613632 model_training_utils.py:505] Train Step: 270/2100 / loss = 1.50537109375 I0421 10:37:41.814551 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 269 and 270 I0421 10:37:42.897418 47076539613632 model_training_utils.py:505] Train Step: 271/2100 / loss = 1.71337890625 I0421 10:37:42.897860 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 270 and 271 I0421 10:37:43.981513 47076539613632 model_training_utils.py:505] Train Step: 272/2100 / loss = 1.6259765625 I0421 10:37:43.981955 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 271 and 272 I0421 10:37:45.067598 47076539613632 model_training_utils.py:505] Train Step: 273/2100 / loss = 1.76611328125 I0421 10:37:45.068038 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 272 and 273 I0421 10:37:46.153645 47076539613632 model_training_utils.py:505] Train Step: 274/2100 / loss = 1.712890625 I0421 10:37:46.154095 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 273 and 274 I0421 10:37:47.238104 47076539613632 model_training_utils.py:505] Train Step: 275/2100 / loss = 2.099853515625 I0421 10:37:47.238545 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 274 and 275 I0421 10:37:48.324071 47076539613632 model_training_utils.py:505] Train Step: 276/2100 / loss = 3.19482421875 I0421 10:37:48.324517 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 275 and 276 I0421 10:37:49.415494 47076539613632 model_training_utils.py:505] Train Step: 277/2100 / loss = 3.095703125 I0421 10:37:49.415925 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.22 examples/second between steps 276 and 277 I0421 10:37:50.500530 47076539613632 model_training_utils.py:505] Train Step: 278/2100 / loss = 2.310546875 I0421 10:37:50.500961 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 277 and 278 I0421 10:37:51.586648 47076539613632 model_training_utils.py:505] Train Step: 279/2100 / loss = 1.547607421875 I0421 10:37:51.587082 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 278 and 279 I0421 10:37:52.674976 47076539613632 model_training_utils.py:505] Train Step: 280/2100 / loss = 1.621826171875 I0421 10:37:52.675422 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.54 examples/second between steps 279 and 280 I0421 10:37:53.760400 47076539613632 model_training_utils.py:505] Train Step: 281/2100 / loss = 1.564208984375 I0421 10:37:53.760834 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 280 and 281 I0421 10:37:54.845687 47076539613632 model_training_utils.py:505] Train Step: 282/2100 / loss = 1.84716796875 I0421 10:37:54.846131 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 281 and 282 I0421 10:37:55.928692 47076539613632 model_training_utils.py:505] Train Step: 283/2100 / loss = 1.88720703125 I0421 10:37:55.929127 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 282 and 283 I0421 10:37:57.012435 47076539613632 model_training_utils.py:505] Train Step: 284/2100 / loss = 1.80908203125 I0421 10:37:57.012863 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 283 and 284 I0421 10:37:58.097065 47076539613632 model_training_utils.py:505] Train Step: 285/2100 / loss = 2.8671875 I0421 10:37:58.097504 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 284 and 285 I0421 10:37:59.180255 47076539613632 model_training_utils.py:505] Train Step: 286/2100 / loss = 2.86083984375 I0421 10:37:59.180702 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 285 and 286 I0421 10:38:00.267987 47076539613632 model_training_utils.py:505] Train Step: 287/2100 / loss = 2.06640625 I0421 10:38:00.268426 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.57 examples/second between steps 286 and 287 I0421 10:38:01.360349 47076539613632 model_training_utils.py:505] Train Step: 288/2100 / loss = 2.633544921875 I0421 10:38:01.360787 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.05 examples/second between steps 287 and 288 I0421 10:38:02.448136 47076539613632 model_training_utils.py:505] Train Step: 289/2100 / loss = 3.0361328125 I0421 10:38:02.448577 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.59 examples/second between steps 288 and 289 I0421 10:38:03.537286 47076539613632 model_training_utils.py:505] Train Step: 290/2100 / loss = 2.70947265625 I0421 10:38:03.537724 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.43 examples/second between steps 289 and 290 I0421 10:38:04.623644 47076539613632 model_training_utils.py:505] Train Step: 291/2100 / loss = 2.33642578125 I0421 10:38:04.624069 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 290 and 291 I0421 10:38:05.707680 47076539613632 model_training_utils.py:505] Train Step: 292/2100 / loss = 1.6201171875 I0421 10:38:05.708124 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 291 and 292 I0421 10:38:06.796768 47076539613632 model_training_utils.py:505] Train Step: 293/2100 / loss = 1.678955078125 I0421 10:38:06.797196 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.45 examples/second between steps 292 and 293 I0421 10:38:07.883083 47076539613632 model_training_utils.py:505] Train Step: 294/2100 / loss = 1.6884765625 I0421 10:38:07.883518 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 293 and 294 I0421 10:38:08.970413 47076539613632 model_training_utils.py:505] Train Step: 295/2100 / loss = 1.765625 I0421 10:38:08.971602 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.54 examples/second between steps 294 and 295 I0421 10:38:10.055443 47076539613632 model_training_utils.py:505] Train Step: 296/2100 / loss = 1.902587890625 I0421 10:38:10.055874 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 295 and 296 I0421 10:38:11.141615 47076539613632 model_training_utils.py:505] Train Step: 297/2100 / loss = 1.730712890625 I0421 10:38:11.142053 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 296 and 297 I0421 10:38:12.225957 47076539613632 model_training_utils.py:505] Train Step: 298/2100 / loss = 1.768310546875 I0421 10:38:12.226391 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 297 and 298 I0421 10:38:13.312524 47076539613632 model_training_utils.py:505] Train Step: 299/2100 / loss = 1.8427734375 I0421 10:38:13.312957 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 298 and 299 I0421 10:38:14.398540 47076539613632 model_training_utils.py:505] Train Step: 300/2100 / loss = 2.079345703125 I0421 10:38:14.398971 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 299 and 300 I0421 10:38:15.484151 47076539613632 model_training_utils.py:505] Train Step: 301/2100 / loss = 1.74462890625 I0421 10:38:15.484595 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 300 and 301 I0421 10:38:16.573475 47076539613632 model_training_utils.py:505] Train Step: 302/2100 / loss = 1.829833984375 I0421 10:38:16.573924 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.40 examples/second between steps 301 and 302 I0421 10:38:17.663332 47076539613632 model_training_utils.py:505] Train Step: 303/2100 / loss = 1.9140625 I0421 10:38:17.663762 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.40 examples/second between steps 302 and 303 I0421 10:38:18.749788 47076539613632 model_training_utils.py:505] Train Step: 304/2100 / loss = 1.567138671875 I0421 10:38:18.750222 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 303 and 304 I0421 10:38:19.835153 47076539613632 model_training_utils.py:505] Train Step: 305/2100 / loss = 1.332275390625 I0421 10:38:19.835597 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 304 and 305 I0421 10:38:20.919651 47076539613632 model_training_utils.py:505] Train Step: 306/2100 / loss = 1.061767578125 I0421 10:38:20.920087 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 305 and 306 I0421 10:38:22.003611 47076539613632 model_training_utils.py:505] Train Step: 307/2100 / loss = 1.541748046875 I0421 10:38:22.004045 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 306 and 307 I0421 10:38:23.087404 47076539613632 model_training_utils.py:505] Train Step: 308/2100 / loss = 1.0927734375 I0421 10:38:23.087837 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 307 and 308 I0421 10:38:24.169559 47076539613632 model_training_utils.py:505] Train Step: 309/2100 / loss = 1.314697265625 I0421 10:38:24.170013 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 308 and 309 I0421 10:38:25.254856 47076539613632 model_training_utils.py:505] Train Step: 310/2100 / loss = 1.45166015625 I0421 10:38:25.255253 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 309 and 310 I0421 10:38:26.341274 47076539613632 model_training_utils.py:505] Train Step: 311/2100 / loss = 1.30419921875 I0421 10:38:26.341703 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 310 and 311 I0421 10:38:27.425168 47076539613632 model_training_utils.py:505] Train Step: 312/2100 / loss = 1.178955078125 I0421 10:38:27.425615 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 311 and 312 I0421 10:38:28.509448 47076539613632 model_training_utils.py:505] Train Step: 313/2100 / loss = 1.134765625 I0421 10:38:28.509877 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 312 and 313 I0421 10:38:29.597739 47076539613632 model_training_utils.py:505] Train Step: 314/2100 / loss = 1.88330078125 I0421 10:38:29.598181 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.51 examples/second between steps 313 and 314 I0421 10:38:30.685404 47076539613632 model_training_utils.py:505] Train Step: 315/2100 / loss = 2.375 I0421 10:38:30.685853 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.57 examples/second between steps 314 and 315 I0421 10:38:31.769310 47076539613632 model_training_utils.py:505] Train Step: 316/2100 / loss = 1.4326171875 I0421 10:38:31.769736 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 315 and 316 I0421 10:38:32.854553 47076539613632 model_training_utils.py:505] Train Step: 317/2100 / loss = 1.6572265625 I0421 10:38:32.854992 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 316 and 317 I0421 10:38:33.939206 47076539613632 model_training_utils.py:505] Train Step: 318/2100 / loss = 1.01025390625 I0421 10:38:33.939652 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 317 and 318 I0421 10:38:35.024049 47076539613632 model_training_utils.py:505] Train Step: 319/2100 / loss = 1.052490234375 I0421 10:38:35.024492 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 318 and 319 I0421 10:38:36.109975 47076539613632 model_training_utils.py:505] Train Step: 320/2100 / loss = 2.75048828125 I0421 10:38:36.110430 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 319 and 320 I0421 10:38:37.200154 47076539613632 model_training_utils.py:505] Train Step: 321/2100 / loss = 2.43212890625 I0421 10:38:37.200596 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.34 examples/second between steps 320 and 321 I0421 10:38:38.286348 47076539613632 model_training_utils.py:505] Train Step: 322/2100 / loss = 1.92431640625 I0421 10:38:38.286773 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 321 and 322 I0421 10:38:39.373057 47076539613632 model_training_utils.py:505] Train Step: 323/2100 / loss = 2.130859375 I0421 10:38:39.373486 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 322 and 323 I0421 10:38:40.463286 47076539613632 model_training_utils.py:505] Train Step: 324/2100 / loss = 1.650390625 I0421 10:38:40.463715 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.29 examples/second between steps 323 and 324 I0421 10:38:41.549247 47076539613632 model_training_utils.py:505] Train Step: 325/2100 / loss = 1.492919921875 I0421 10:38:41.549694 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 324 and 325 I0421 10:38:42.636014 47076539613632 model_training_utils.py:505] Train Step: 326/2100 / loss = 1.71484375 I0421 10:38:42.636457 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 325 and 326 I0421 10:38:43.724728 47076539613632 model_training_utils.py:505] Train Step: 327/2100 / loss = 2.143310546875 I0421 10:38:43.725167 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.47 examples/second between steps 326 and 327 I0421 10:38:44.818813 47076539613632 model_training_utils.py:505] Train Step: 328/2100 / loss = 1.66552734375 I0421 10:38:44.819251 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.91 examples/second between steps 327 and 328 I0421 10:38:45.905325 47076539613632 model_training_utils.py:505] Train Step: 329/2100 / loss = 1.24951171875 I0421 10:38:45.905770 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 328 and 329 I0421 10:38:46.994952 47076539613632 model_training_utils.py:505] Train Step: 330/2100 / loss = 1.25537109375 I0421 10:38:46.995393 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.38 examples/second between steps 329 and 330 I0421 10:38:48.085377 47076539613632 model_training_utils.py:505] Train Step: 331/2100 / loss = 1.53857421875 I0421 10:38:48.085815 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.28 examples/second between steps 330 and 331 I0421 10:38:49.177045 47076539613632 model_training_utils.py:505] Train Step: 332/2100 / loss = 1.911865234375 I0421 10:38:49.177488 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.15 examples/second between steps 331 and 332 I0421 10:38:50.262437 47076539613632 model_training_utils.py:505] Train Step: 333/2100 / loss = 2.67626953125 I0421 10:38:50.262840 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 332 and 333 I0421 10:38:51.349210 47076539613632 model_training_utils.py:505] Train Step: 334/2100 / loss = 2.0205078125 I0421 10:38:51.349642 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.70 examples/second between steps 333 and 334 I0421 10:38:52.443220 47076539613632 model_training_utils.py:505] Train Step: 335/2100 / loss = 2.02734375 I0421 10:38:52.443661 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.89 examples/second between steps 334 and 335 I0421 10:38:53.528214 47076539613632 model_training_utils.py:505] Train Step: 336/2100 / loss = 1.795166015625 I0421 10:38:53.528657 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 335 and 336 I0421 10:38:54.613633 47076539613632 model_training_utils.py:505] Train Step: 337/2100 / loss = 1.3046875 I0421 10:38:54.614074 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 336 and 337 I0421 10:38:55.699162 47076539613632 model_training_utils.py:505] Train Step: 338/2100 / loss = 1.416015625 I0421 10:38:55.699598 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 337 and 338 I0421 10:38:56.785155 47076539613632 model_training_utils.py:505] Train Step: 339/2100 / loss = 1.08447265625 I0421 10:38:56.785596 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 338 and 339 I0421 10:38:57.872408 47076539613632 model_training_utils.py:505] Train Step: 340/2100 / loss = 1.51123046875 I0421 10:38:57.872846 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.64 examples/second between steps 339 and 340 I0421 10:38:58.963851 47076539613632 model_training_utils.py:505] Train Step: 341/2100 / loss = 1.9267578125 I0421 10:38:58.964295 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.18 examples/second between steps 340 and 341 I0421 10:39:00.056198 47076539613632 model_training_utils.py:505] Train Step: 342/2100 / loss = 1.530029296875 I0421 10:39:00.056620 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.15 examples/second between steps 341 and 342 I0421 10:39:01.146629 47076539613632 model_training_utils.py:505] Train Step: 343/2100 / loss = 2.79833984375 I0421 10:39:01.147047 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.31 examples/second between steps 342 and 343 I0421 10:39:02.240182 47076539613632 model_training_utils.py:505] Train Step: 344/2100 / loss = 2.035888671875 I0421 10:39:02.240573 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.95 examples/second between steps 343 and 344 I0421 10:39:03.334143 47076539613632 model_training_utils.py:505] Train Step: 345/2100 / loss = 1.9658203125 I0421 10:39:03.334534 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.92 examples/second between steps 344 and 345 I0421 10:39:04.428382 47076539613632 model_training_utils.py:505] Train Step: 346/2100 / loss = 1.9287109375 I0421 10:39:04.428781 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.86 examples/second between steps 345 and 346 I0421 10:39:05.516049 47076539613632 model_training_utils.py:505] Train Step: 347/2100 / loss = 1.83935546875 I0421 10:39:05.516443 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.61 examples/second between steps 346 and 347 I0421 10:39:06.603738 47076539613632 model_training_utils.py:505] Train Step: 348/2100 / loss = 1.912353515625 I0421 10:39:06.604121 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.59 examples/second between steps 347 and 348 I0421 10:39:07.682903 47076539613632 model_training_utils.py:505] Train Step: 349/2100 / loss = 1.7158203125 I0421 10:39:07.683290 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.56 examples/second between steps 348 and 349 I0421 10:39:08.772351 47076539613632 model_training_utils.py:505] Train Step: 350/2100 / loss = 1.451416015625 I0421 10:39:08.772736 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.41 examples/second between steps 349 and 350 I0421 10:39:09.859643 47076539613632 model_training_utils.py:505] Train Step: 351/2100 / loss = 1.799072265625 I0421 10:39:09.860026 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.64 examples/second between steps 350 and 351 I0421 10:39:10.945577 47076539613632 model_training_utils.py:505] Train Step: 352/2100 / loss = 1.81591796875 I0421 10:39:10.945961 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 351 and 352 I0421 10:39:12.032278 47076539613632 model_training_utils.py:505] Train Step: 353/2100 / loss = 1.724609375 I0421 10:39:12.032680 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 352 and 353 I0421 10:39:13.115339 47076539613632 model_training_utils.py:505] Train Step: 354/2100 / loss = 1.543212890625 I0421 10:39:13.115725 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 353 and 354 I0421 10:39:14.202596 47076539613632 model_training_utils.py:505] Train Step: 355/2100 / loss = 1.6552734375 I0421 10:39:14.202981 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 354 and 355 I0421 10:39:15.287914 47076539613632 model_training_utils.py:505] Train Step: 356/2100 / loss = 1.632080078125 I0421 10:39:15.288303 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 355 and 356 I0421 10:39:16.373405 47076539613632 model_training_utils.py:505] Train Step: 357/2100 / loss = 1.599853515625 I0421 10:39:16.373788 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 356 and 357 I0421 10:39:17.459241 47076539613632 model_training_utils.py:505] Train Step: 358/2100 / loss = 1.43115234375 I0421 10:39:17.459629 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 357 and 358 I0421 10:39:18.545299 47076539613632 model_training_utils.py:505] Train Step: 359/2100 / loss = 2.122314453125 I0421 10:39:18.545689 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 358 and 359 I0421 10:39:19.630056 47076539613632 model_training_utils.py:505] Train Step: 360/2100 / loss = 2.619140625 I0421 10:39:19.630446 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 359 and 360 I0421 10:39:20.714480 47076539613632 model_training_utils.py:505] Train Step: 361/2100 / loss = 1.74169921875 I0421 10:39:20.714870 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 360 and 361 I0421 10:39:21.796844 47076539613632 model_training_utils.py:505] Train Step: 362/2100 / loss = 1.75341796875 I0421 10:39:21.797224 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 361 and 362 I0421 10:39:22.880102 47076539613632 model_training_utils.py:505] Train Step: 363/2100 / loss = 1.482421875 I0421 10:39:22.880493 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 362 and 363 I0421 10:39:23.963794 47076539613632 model_training_utils.py:505] Train Step: 364/2100 / loss = 2.209228515625 I0421 10:39:23.964174 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 363 and 364 I0421 10:39:25.050236 47076539613632 model_training_utils.py:505] Train Step: 365/2100 / loss = 2.3818359375 I0421 10:39:25.050626 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 364 and 365 I0421 10:39:26.133832 47076539613632 model_training_utils.py:505] Train Step: 366/2100 / loss = 2.22509765625 I0421 10:39:26.134209 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 365 and 366 I0421 10:39:27.217152 47076539613632 model_training_utils.py:505] Train Step: 367/2100 / loss = 1.7421875 I0421 10:39:27.217537 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 366 and 367 I0421 10:39:28.301435 47076539613632 model_training_utils.py:505] Train Step: 368/2100 / loss = 1.29541015625 I0421 10:39:28.301821 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 367 and 368 I0421 10:39:29.385930 47076539613632 model_training_utils.py:505] Train Step: 369/2100 / loss = 1.7861328125 I0421 10:39:29.386322 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 368 and 369 I0421 10:39:30.470473 47076539613632 model_training_utils.py:505] Train Step: 370/2100 / loss = 2.54150390625 I0421 10:39:30.470860 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 369 and 370 I0421 10:39:31.557722 47076539613632 model_training_utils.py:505] Train Step: 371/2100 / loss = 2.1982421875 I0421 10:39:31.558102 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.61 examples/second between steps 370 and 371 I0421 10:39:32.643596 47076539613632 model_training_utils.py:505] Train Step: 372/2100 / loss = 1.394287109375 I0421 10:39:32.643977 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 371 and 372 I0421 10:39:33.723816 47076539613632 model_training_utils.py:505] Train Step: 373/2100 / loss = 1.514404296875 I0421 10:39:33.724196 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.39 examples/second between steps 372 and 373 I0421 10:39:34.809803 47076539613632 model_training_utils.py:505] Train Step: 374/2100 / loss = 1.431396484375 I0421 10:39:34.810189 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 373 and 374 I0421 10:39:35.897136 47076539613632 model_training_utils.py:505] Train Step: 375/2100 / loss = 1.524658203125 I0421 10:39:35.897531 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 374 and 375 I0421 10:39:36.991240 47076539613632 model_training_utils.py:505] Train Step: 376/2100 / loss = 1.32470703125 I0421 10:39:36.991638 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.88 examples/second between steps 375 and 376 I0421 10:39:38.089576 47076539613632 model_training_utils.py:505] Train Step: 377/2100 / loss = 1.681884765625 I0421 10:39:38.089957 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.43 examples/second between steps 376 and 377 I0421 10:39:39.180543 47076539613632 model_training_utils.py:505] Train Step: 378/2100 / loss = 2.443359375 I0421 10:39:39.180927 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.23 examples/second between steps 377 and 378 I0421 10:39:40.274457 47076539613632 model_training_utils.py:505] Train Step: 379/2100 / loss = 2.86328125 I0421 10:39:40.274840 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.91 examples/second between steps 378 and 379 I0421 10:39:41.368330 47076539613632 model_training_utils.py:505] Train Step: 380/2100 / loss = 3.4267578125 I0421 10:39:41.368717 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.01 examples/second between steps 379 and 380 I0421 10:39:42.454418 47076539613632 model_training_utils.py:505] Train Step: 381/2100 / loss = 2.82666015625 I0421 10:39:42.454809 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 380 and 381 I0421 10:39:43.540168 47076539613632 model_training_utils.py:505] Train Step: 382/2100 / loss = 1.896728515625 I0421 10:39:43.540552 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 381 and 382 I0421 10:39:44.628319 47076539613632 model_training_utils.py:505] Train Step: 383/2100 / loss = 1.648193359375 I0421 10:39:44.628702 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.54 examples/second between steps 382 and 383 I0421 10:39:45.721182 47076539613632 model_training_utils.py:505] Train Step: 384/2100 / loss = 1.7333984375 I0421 10:39:45.721571 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.03 examples/second between steps 383 and 384 I0421 10:39:46.810573 47076539613632 model_training_utils.py:505] Train Step: 385/2100 / loss = 1.60693359375 I0421 10:39:46.810966 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.41 examples/second between steps 384 and 385 I0421 10:39:47.896821 47076539613632 model_training_utils.py:505] Train Step: 386/2100 / loss = 1.568115234375 I0421 10:39:47.897206 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 385 and 386 I0421 10:39:48.984844 47076539613632 model_training_utils.py:505] Train Step: 387/2100 / loss = 1.3328857421875 I0421 10:39:48.985229 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 386 and 387 I0421 10:39:50.073522 47076539613632 model_training_utils.py:505] Train Step: 388/2100 / loss = 1.654052734375 I0421 10:39:50.073906 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.45 examples/second between steps 387 and 388 I0421 10:39:51.155300 47076539613632 model_training_utils.py:505] Train Step: 389/2100 / loss = 2.49462890625 I0421 10:39:51.155677 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 388 and 389 I0421 10:39:52.238724 47076539613632 model_training_utils.py:505] Train Step: 390/2100 / loss = 2.3427734375 I0421 10:39:52.239104 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 389 and 390 I0421 10:39:53.322158 47076539613632 model_training_utils.py:505] Train Step: 391/2100 / loss = 2.06396484375 I0421 10:39:53.322549 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 390 and 391 I0421 10:39:54.407061 47076539613632 model_training_utils.py:505] Train Step: 392/2100 / loss = 1.482666015625 I0421 10:39:54.407454 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 391 and 392 I0421 10:39:55.491780 47076539613632 model_training_utils.py:505] Train Step: 393/2100 / loss = 1.1932373046875 I0421 10:39:55.492161 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 392 and 393 I0421 10:39:56.578824 47076539613632 model_training_utils.py:505] Train Step: 394/2100 / loss = 1.1162109375 I0421 10:39:56.579201 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 393 and 394 I0421 10:39:57.666678 47076539613632 model_training_utils.py:505] Train Step: 395/2100 / loss = 1.519775390625 I0421 10:39:57.667057 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.55 examples/second between steps 394 and 395 I0421 10:39:58.751467 47076539613632 model_training_utils.py:505] Train Step: 396/2100 / loss = 1.77734375 I0421 10:39:58.751851 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 395 and 396 I0421 10:39:59.835806 47076539613632 model_training_utils.py:505] Train Step: 397/2100 / loss = 1.5361328125 I0421 10:39:59.836192 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 396 and 397 I0421 10:40:00.922264 47076539613632 model_training_utils.py:505] Train Step: 398/2100 / loss = 1.4033203125 I0421 10:40:00.922650 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 397 and 398 I0421 10:40:02.009158 47076539613632 model_training_utils.py:505] Train Step: 399/2100 / loss = 1.443115234375 I0421 10:40:02.009553 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 398 and 399 I0421 10:40:03.096000 47076539613632 model_training_utils.py:505] Train Step: 400/2100 / loss = 1.2431640625 I0421 10:40:03.096395 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 399 and 400 I0421 10:40:04.183169 47076539613632 model_training_utils.py:505] Train Step: 401/2100 / loss = 1.28125 I0421 10:40:04.183565 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.62 examples/second between steps 400 and 401 I0421 10:40:05.269026 47076539613632 model_training_utils.py:505] Train Step: 402/2100 / loss = 1.3994140625 I0421 10:40:05.269415 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 401 and 402 I0421 10:40:06.350570 47076539613632 model_training_utils.py:505] Train Step: 403/2100 / loss = 1.551025390625 I0421 10:40:06.350955 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 402 and 403 I0421 10:40:07.434485 47076539613632 model_training_utils.py:505] Train Step: 404/2100 / loss = 1.633056640625 I0421 10:40:07.434868 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 403 and 404 I0421 10:40:08.518422 47076539613632 model_training_utils.py:505] Train Step: 405/2100 / loss = 1.4739990234375 I0421 10:40:08.518802 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 404 and 405 I0421 10:40:09.604782 47076539613632 model_training_utils.py:505] Train Step: 406/2100 / loss = 2.21044921875 I0421 10:40:09.605165 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 405 and 406 I0421 10:40:10.690788 47076539613632 model_training_utils.py:505] Train Step: 407/2100 / loss = 1.629638671875 I0421 10:40:10.691168 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 406 and 407 I0421 10:40:11.781507 47076539613632 model_training_utils.py:505] Train Step: 408/2100 / loss = 1.3699951171875 I0421 10:40:11.781885 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.26 examples/second between steps 407 and 408 I0421 10:40:12.865730 47076539613632 model_training_utils.py:505] Train Step: 409/2100 / loss = 1.2578125 I0421 10:40:12.866117 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 408 and 409 I0421 10:40:13.949761 47076539613632 model_training_utils.py:505] Train Step: 410/2100 / loss = 1.537109375 I0421 10:40:13.950145 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 409 and 410 I0421 10:40:15.041076 47076539613632 model_training_utils.py:505] Train Step: 411/2100 / loss = 2.052734375 I0421 10:40:15.041463 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.16 examples/second between steps 410 and 411 I0421 10:40:16.130613 47076539613632 model_training_utils.py:505] Train Step: 412/2100 / loss = 2.089599609375 I0421 10:40:16.130990 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.37 examples/second between steps 411 and 412 I0421 10:40:17.214931 47076539613632 model_training_utils.py:505] Train Step: 413/2100 / loss = 1.66845703125 I0421 10:40:17.215322 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 412 and 413 I0421 10:40:18.301891 47076539613632 model_training_utils.py:505] Train Step: 414/2100 / loss = 1.67333984375 I0421 10:40:18.302275 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 413 and 414 I0421 10:40:19.388364 47076539613632 model_training_utils.py:505] Train Step: 415/2100 / loss = 1.22216796875 I0421 10:40:19.388756 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 414 and 415 I0421 10:40:20.474512 47076539613632 model_training_utils.py:505] Train Step: 416/2100 / loss = 1.2392578125 I0421 10:40:20.474895 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 415 and 416 I0421 10:40:21.562211 47076539613632 model_training_utils.py:505] Train Step: 417/2100 / loss = 1.530517578125 I0421 10:40:21.562602 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 416 and 417 I0421 10:40:22.652662 47076539613632 model_training_utils.py:505] Train Step: 418/2100 / loss = 1.85205078125 I0421 10:40:22.653043 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.30 examples/second between steps 417 and 418 I0421 10:40:23.742236 47076539613632 model_training_utils.py:505] Train Step: 419/2100 / loss = 1.720458984375 I0421 10:40:23.742627 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.37 examples/second between steps 418 and 419 I0421 10:40:24.828571 47076539613632 model_training_utils.py:505] Train Step: 420/2100 / loss = 1.53173828125 I0421 10:40:24.828954 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 419 and 420 I0421 10:40:25.915418 47076539613632 model_training_utils.py:505] Train Step: 421/2100 / loss = 1.6787109375 I0421 10:40:25.915802 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 420 and 421 I0421 10:40:27.001512 47076539613632 model_training_utils.py:505] Train Step: 422/2100 / loss = 2.03076171875 I0421 10:40:27.001892 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 421 and 422 I0421 10:40:28.089389 47076539613632 model_training_utils.py:505] Train Step: 423/2100 / loss = 1.39306640625 I0421 10:40:28.089783 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.56 examples/second between steps 422 and 423 I0421 10:40:29.177295 47076539613632 model_training_utils.py:505] Train Step: 424/2100 / loss = 1.773681640625 I0421 10:40:29.177679 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.57 examples/second between steps 423 and 424 I0421 10:40:30.258994 47076539613632 model_training_utils.py:505] Train Step: 425/2100 / loss = 2.32958984375 I0421 10:40:30.259385 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.25 examples/second between steps 424 and 425 I0421 10:40:31.343122 47076539613632 model_training_utils.py:505] Train Step: 426/2100 / loss = 2.0205078125 I0421 10:40:31.343528 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 425 and 426 I0421 10:40:32.426244 47076539613632 model_training_utils.py:505] Train Step: 427/2100 / loss = 2.4775390625 I0421 10:40:32.426633 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 426 and 427 I0421 10:40:33.510426 47076539613632 model_training_utils.py:505] Train Step: 428/2100 / loss = 2.147705078125 I0421 10:40:33.510809 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 427 and 428 I0421 10:40:34.596785 47076539613632 model_training_utils.py:505] Train Step: 429/2100 / loss = 2.05615234375 I0421 10:40:34.597170 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 428 and 429 I0421 10:40:35.683142 47076539613632 model_training_utils.py:505] Train Step: 430/2100 / loss = 1.56982421875 I0421 10:40:35.683535 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 429 and 430 I0421 10:40:36.767720 47076539613632 model_training_utils.py:505] Train Step: 431/2100 / loss = 1.450927734375 I0421 10:40:36.768118 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 430 and 431 I0421 10:40:37.852011 47076539613632 model_training_utils.py:505] Train Step: 432/2100 / loss = 1.63720703125 I0421 10:40:37.852403 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 431 and 432 I0421 10:40:38.936460 47076539613632 model_training_utils.py:505] Train Step: 433/2100 / loss = 1.48095703125 I0421 10:40:38.936843 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 432 and 433 I0421 10:40:40.021685 47076539613632 model_training_utils.py:505] Train Step: 434/2100 / loss = 1.731689453125 I0421 10:40:40.022075 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 433 and 434 I0421 10:40:41.109047 47076539613632 model_training_utils.py:505] Train Step: 435/2100 / loss = 1.755126953125 I0421 10:40:41.109439 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.62 examples/second between steps 434 and 435 I0421 10:40:42.195044 47076539613632 model_training_utils.py:505] Train Step: 436/2100 / loss = 1.84375 I0421 10:40:42.195464 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 435 and 436 I0421 10:40:43.283028 47076539613632 model_training_utils.py:505] Train Step: 437/2100 / loss = 1.78271484375 I0421 10:40:43.283417 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 436 and 437 I0421 10:40:44.370090 47076539613632 model_training_utils.py:505] Train Step: 438/2100 / loss = 1.63916015625 I0421 10:40:44.370490 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 437 and 438 I0421 10:40:45.454013 47076539613632 model_training_utils.py:505] Train Step: 439/2100 / loss = 1.396240234375 I0421 10:40:45.454409 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 438 and 439 I0421 10:40:46.541704 47076539613632 model_training_utils.py:505] Train Step: 440/2100 / loss = 1.58251953125 I0421 10:40:46.542088 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 439 and 440 I0421 10:40:47.627062 47076539613632 model_training_utils.py:505] Train Step: 441/2100 / loss = 2.65966796875 I0421 10:40:47.627456 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 440 and 441 I0421 10:40:48.711026 47076539613632 model_training_utils.py:505] Train Step: 442/2100 / loss = 2.497314453125 I0421 10:40:48.711419 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 441 and 442 I0421 10:40:49.795244 47076539613632 model_training_utils.py:505] Train Step: 443/2100 / loss = 1.43408203125 I0421 10:40:49.795637 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 442 and 443 I0421 10:40:50.883152 47076539613632 model_training_utils.py:505] Train Step: 444/2100 / loss = 1.95556640625 I0421 10:40:50.883540 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.55 examples/second between steps 443 and 444 I0421 10:40:51.968538 47076539613632 model_training_utils.py:505] Train Step: 445/2100 / loss = 1.79736328125 I0421 10:40:51.968920 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 444 and 445 I0421 10:40:53.053330 47076539613632 model_training_utils.py:505] Train Step: 446/2100 / loss = 1.515380859375 I0421 10:40:53.053714 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 445 and 446 I0421 10:40:54.138927 47076539613632 model_training_utils.py:505] Train Step: 447/2100 / loss = 1.1475830078125 I0421 10:40:54.139338 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 446 and 447 I0421 10:40:55.225168 47076539613632 model_training_utils.py:505] Train Step: 448/2100 / loss = 1.07421875 I0421 10:40:55.225553 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 447 and 448 I0421 10:40:56.307857 47076539613632 model_training_utils.py:505] Train Step: 449/2100 / loss = 1.56982421875 I0421 10:40:56.308241 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 448 and 449 I0421 10:40:57.394321 47076539613632 model_training_utils.py:505] Train Step: 450/2100 / loss = 2.082763671875 I0421 10:40:57.394710 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 449 and 450 I0421 10:40:58.480298 47076539613632 model_training_utils.py:505] Train Step: 451/2100 / loss = 2.02392578125 I0421 10:40:58.480684 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 450 and 451 I0421 10:40:59.567233 47076539613632 model_training_utils.py:505] Train Step: 452/2100 / loss = 1.502197265625 I0421 10:40:59.567626 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 451 and 452 I0421 10:41:00.653912 47076539613632 model_training_utils.py:505] Train Step: 453/2100 / loss = 1.94921875 I0421 10:41:00.654314 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.70 examples/second between steps 452 and 453 I0421 10:41:01.740939 47076539613632 model_training_utils.py:505] Train Step: 454/2100 / loss = 2.2197265625 I0421 10:41:01.741325 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 453 and 454 I0421 10:41:02.825470 47076539613632 model_training_utils.py:505] Train Step: 455/2100 / loss = 1.76318359375 I0421 10:41:02.825851 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 454 and 455 I0421 10:41:03.915193 47076539613632 model_training_utils.py:505] Train Step: 456/2100 / loss = 1.8232421875 I0421 10:41:03.915581 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.37 examples/second between steps 455 and 456 I0421 10:41:05.005789 47076539613632 model_training_utils.py:505] Train Step: 457/2100 / loss = 2.4482421875 I0421 10:41:05.006172 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.27 examples/second between steps 456 and 457 I0421 10:41:06.088326 47076539613632 model_training_utils.py:505] Train Step: 458/2100 / loss = 2.48046875 I0421 10:41:06.088709 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 457 and 458 I0421 10:41:07.174426 47076539613632 model_training_utils.py:505] Train Step: 459/2100 / loss = 2.156005859375 I0421 10:41:07.174810 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 458 and 459 I0421 10:41:08.261306 47076539613632 model_training_utils.py:505] Train Step: 460/2100 / loss = 2.24365234375 I0421 10:41:08.261686 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 459 and 460 I0421 10:41:09.348405 47076539613632 model_training_utils.py:505] Train Step: 461/2100 / loss = 2.2890625 I0421 10:41:09.348790 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.64 examples/second between steps 460 and 461 I0421 10:41:10.434391 47076539613632 model_training_utils.py:505] Train Step: 462/2100 / loss = 1.622802734375 I0421 10:41:10.434780 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 461 and 462 I0421 10:41:11.519661 47076539613632 model_training_utils.py:505] Train Step: 463/2100 / loss = 1.44677734375 I0421 10:41:11.520041 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 462 and 463 I0421 10:41:12.603363 47076539613632 model_training_utils.py:505] Train Step: 464/2100 / loss = 1.601318359375 I0421 10:41:12.603746 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 463 and 464 I0421 10:41:13.687149 47076539613632 model_training_utils.py:505] Train Step: 465/2100 / loss = 1.8955078125 I0421 10:41:13.687552 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 464 and 465 I0421 10:41:14.774652 47076539613632 model_training_utils.py:505] Train Step: 466/2100 / loss = 1.71142578125 I0421 10:41:14.775038 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.57 examples/second between steps 465 and 466 I0421 10:41:15.859898 47076539613632 model_training_utils.py:505] Train Step: 467/2100 / loss = 1.6357421875 I0421 10:41:15.860289 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 466 and 467 I0421 10:41:16.945927 47076539613632 model_training_utils.py:505] Train Step: 468/2100 / loss = 1.62646484375 I0421 10:41:16.946314 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 467 and 468 I0421 10:41:18.035357 47076539613632 model_training_utils.py:505] Train Step: 469/2100 / loss = 1.7939453125 I0421 10:41:18.035739 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.37 examples/second between steps 468 and 469 I0421 10:41:19.118366 47076539613632 model_training_utils.py:505] Train Step: 470/2100 / loss = 2.029052734375 I0421 10:41:19.118748 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 469 and 470 I0421 10:41:20.203003 47076539613632 model_training_utils.py:505] Train Step: 471/2100 / loss = 1.457275390625 I0421 10:41:20.203394 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 470 and 471 I0421 10:41:21.284784 47076539613632 model_training_utils.py:505] Train Step: 472/2100 / loss = 1.226318359375 I0421 10:41:21.285162 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 471 and 472 I0421 10:41:22.374432 47076539613632 model_training_utils.py:505] Train Step: 473/2100 / loss = 1.537109375 I0421 10:41:22.374820 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.37 examples/second between steps 472 and 473 I0421 10:41:23.457965 47076539613632 model_training_utils.py:505] Train Step: 474/2100 / loss = 1.295166015625 I0421 10:41:23.458356 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 473 and 474 I0421 10:41:24.541465 47076539613632 model_training_utils.py:505] Train Step: 475/2100 / loss = 1.4853515625 I0421 10:41:24.541851 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 474 and 475 I0421 10:41:25.626637 47076539613632 model_training_utils.py:505] Train Step: 476/2100 / loss = 1.522216796875 I0421 10:41:25.627016 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 475 and 476 I0421 10:41:26.714933 47076539613632 model_training_utils.py:505] Train Step: 477/2100 / loss = 1.37548828125 I0421 10:41:26.715322 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.52 examples/second between steps 476 and 477 I0421 10:41:27.804724 47076539613632 model_training_utils.py:505] Train Step: 478/2100 / loss = 1.45361328125 I0421 10:41:27.805107 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.37 examples/second between steps 477 and 478 I0421 10:41:28.890561 47076539613632 model_training_utils.py:505] Train Step: 479/2100 / loss = 1.341796875 I0421 10:41:28.890941 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 478 and 479 I0421 10:41:29.976774 47076539613632 model_training_utils.py:505] Train Step: 480/2100 / loss = 1.663818359375 I0421 10:41:29.977171 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 479 and 480 I0421 10:41:31.065073 47076539613632 model_training_utils.py:505] Train Step: 481/2100 / loss = 1.284423828125 I0421 10:41:31.065465 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.51 examples/second between steps 480 and 481 I0421 10:41:32.152190 47076539613632 model_training_utils.py:505] Train Step: 482/2100 / loss = 0.9088134765625 I0421 10:41:32.152574 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.62 examples/second between steps 481 and 482 I0421 10:41:33.235312 47076539613632 model_training_utils.py:505] Train Step: 483/2100 / loss = 1.0806884765625 I0421 10:41:33.235697 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 482 and 483 I0421 10:41:34.321370 47076539613632 model_training_utils.py:505] Train Step: 484/2100 / loss = 0.9891357421875 I0421 10:41:34.321749 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 483 and 484 I0421 10:41:35.410709 47076539613632 model_training_utils.py:505] Train Step: 485/2100 / loss = 1.046630859375 I0421 10:41:35.411093 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.38 examples/second between steps 484 and 485 I0421 10:41:36.499203 47076539613632 model_training_utils.py:505] Train Step: 486/2100 / loss = 1.2412109375 I0421 10:41:36.499592 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.58 examples/second between steps 485 and 486 I0421 10:41:37.582301 47076539613632 model_training_utils.py:505] Train Step: 487/2100 / loss = 1.21826171875 I0421 10:41:37.582686 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 486 and 487 I0421 10:41:38.668452 47076539613632 model_training_utils.py:505] Train Step: 488/2100 / loss = 1.4439697265625 I0421 10:41:38.668830 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 487 and 488 I0421 10:41:39.752951 47076539613632 model_training_utils.py:505] Train Step: 489/2100 / loss = 1.16015625 I0421 10:41:39.753344 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 488 and 489 I0421 10:41:40.838794 47076539613632 model_training_utils.py:505] Train Step: 490/2100 / loss = 1.163818359375 I0421 10:41:40.839175 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 489 and 490 I0421 10:41:41.923044 47076539613632 model_training_utils.py:505] Train Step: 491/2100 / loss = 1.764892578125 I0421 10:41:41.923436 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 490 and 491 I0421 10:41:43.011649 47076539613632 model_training_utils.py:505] Train Step: 492/2100 / loss = 1.787353515625 I0421 10:41:43.012033 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.49 examples/second between steps 491 and 492 I0421 10:41:44.093919 47076539613632 model_training_utils.py:505] Train Step: 493/2100 / loss = 1.35009765625 I0421 10:41:44.094312 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 492 and 493 I0421 10:41:45.176827 47076539613632 model_training_utils.py:505] Train Step: 494/2100 / loss = 1.40380859375 I0421 10:41:45.177234 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 493 and 494 I0421 10:41:46.261704 47076539613632 model_training_utils.py:505] Train Step: 495/2100 / loss = 1.681884765625 I0421 10:41:46.262089 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 494 and 495 I0421 10:41:47.345225 47076539613632 model_training_utils.py:505] Train Step: 496/2100 / loss = 1.380615234375 I0421 10:41:47.345613 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 495 and 496 I0421 10:41:48.427760 47076539613632 model_training_utils.py:505] Train Step: 497/2100 / loss = 1.53564453125 I0421 10:41:48.428156 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 496 and 497 I0421 10:41:49.514214 47076539613632 model_training_utils.py:505] Train Step: 498/2100 / loss = 1.528564453125 I0421 10:41:49.514605 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.70 examples/second between steps 497 and 498 I0421 10:41:50.599313 47076539613632 model_training_utils.py:505] Train Step: 499/2100 / loss = 1.62353515625 I0421 10:41:50.599693 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 498 and 499 I0421 10:41:51.684388 47076539613632 model_training_utils.py:505] Train Step: 500/2100 / loss = 1.51904296875 I0421 10:41:51.684772 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 499 and 500 I0421 10:41:52.772576 47076539613632 model_training_utils.py:505] Train Step: 501/2100 / loss = 1.682861328125 I0421 10:41:52.772957 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.54 examples/second between steps 500 and 501 I0421 10:41:53.866122 47076539613632 model_training_utils.py:505] Train Step: 502/2100 / loss = 1.630126953125 I0421 10:41:53.866521 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.95 examples/second between steps 501 and 502 I0421 10:41:54.951194 47076539613632 model_training_utils.py:505] Train Step: 503/2100 / loss = 2.096435546875 I0421 10:41:54.951582 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 502 and 503 I0421 10:41:56.035248 47076539613632 model_training_utils.py:505] Train Step: 504/2100 / loss = 1.791259765625 I0421 10:41:56.035640 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 503 and 504 I0421 10:41:57.117562 47076539613632 model_training_utils.py:505] Train Step: 505/2100 / loss = 1.6337890625 I0421 10:41:57.117945 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 504 and 505 I0421 10:41:58.201441 47076539613632 model_training_utils.py:505] Train Step: 506/2100 / loss = 1.74609375 I0421 10:41:58.201822 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 505 and 506 I0421 10:41:59.286799 47076539613632 model_training_utils.py:505] Train Step: 507/2100 / loss = 1.642333984375 I0421 10:41:59.287182 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 506 and 507 I0421 10:42:00.373651 47076539613632 model_training_utils.py:505] Train Step: 508/2100 / loss = 1.537353515625 I0421 10:42:00.374091 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 507 and 508 I0421 10:42:01.459945 47076539613632 model_training_utils.py:505] Train Step: 509/2100 / loss = 1.1630859375 I0421 10:42:01.460398 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 508 and 509 I0421 10:42:02.544161 47076539613632 model_training_utils.py:505] Train Step: 510/2100 / loss = 1.238037109375 I0421 10:42:02.544624 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 509 and 510 I0421 10:42:03.631264 47076539613632 model_training_utils.py:505] Train Step: 511/2100 / loss = 1.2734375 I0421 10:42:03.631710 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 510 and 511 I0421 10:42:04.718500 47076539613632 model_training_utils.py:505] Train Step: 512/2100 / loss = 1.3798828125 I0421 10:42:04.718942 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 511 and 512 I0421 10:42:05.808909 47076539613632 model_training_utils.py:505] Train Step: 513/2100 / loss = 1.3388671875 I0421 10:42:05.809363 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.30 examples/second between steps 512 and 513 I0421 10:42:06.898506 47076539613632 model_training_utils.py:505] Train Step: 514/2100 / loss = 1.74755859375 I0421 10:42:06.898949 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.41 examples/second between steps 513 and 514 I0421 10:42:07.988959 47076539613632 model_training_utils.py:505] Train Step: 515/2100 / loss = 1.69580078125 I0421 10:42:07.989410 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.31 examples/second between steps 514 and 515 I0421 10:42:09.075988 47076539613632 model_training_utils.py:505] Train Step: 516/2100 / loss = 1.3037109375 I0421 10:42:09.076441 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.69 examples/second between steps 515 and 516 I0421 10:42:10.161685 47076539613632 model_training_utils.py:505] Train Step: 517/2100 / loss = 1.4912109375 I0421 10:42:10.162125 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 516 and 517 I0421 10:42:11.248133 47076539613632 model_training_utils.py:505] Train Step: 518/2100 / loss = 1.3154296875 I0421 10:42:11.248584 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 517 and 518 I0421 10:42:12.333856 47076539613632 model_training_utils.py:505] Train Step: 519/2100 / loss = 1.41162109375 I0421 10:42:12.334307 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 518 and 519 I0421 10:42:13.418418 47076539613632 model_training_utils.py:505] Train Step: 520/2100 / loss = 1.294921875 I0421 10:42:13.418848 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 519 and 520 I0421 10:42:14.501243 47076539613632 model_training_utils.py:505] Train Step: 521/2100 / loss = 1.2039794921875 I0421 10:42:14.501670 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 520 and 521 I0421 10:42:15.587245 47076539613632 model_training_utils.py:505] Train Step: 522/2100 / loss = 0.9847412109375 I0421 10:42:15.587685 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 521 and 522 I0421 10:42:16.668939 47076539613632 model_training_utils.py:505] Train Step: 523/2100 / loss = 0.931640625 I0421 10:42:16.669378 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 522 and 523 I0421 10:42:17.756553 47076539613632 model_training_utils.py:505] Train Step: 524/2100 / loss = 1.0362548828125 I0421 10:42:17.756986 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 523 and 524 I0421 10:42:18.844296 47076539613632 model_training_utils.py:505] Train Step: 525/2100 / loss = 0.9427490234375 I0421 10:42:18.844731 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.59 examples/second between steps 524 and 525 I0421 10:42:19.924770 47076539613632 model_training_utils.py:505] Train Step: 526/2100 / loss = 1.274658203125 I0421 10:42:19.925209 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.37 examples/second between steps 525 and 526 I0421 10:42:21.013316 47076539613632 model_training_utils.py:505] Train Step: 527/2100 / loss = 1.764404296875 I0421 10:42:21.013760 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.49 examples/second between steps 526 and 527 I0421 10:42:22.097192 47076539613632 model_training_utils.py:505] Train Step: 528/2100 / loss = 1.70361328125 I0421 10:42:22.097624 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 527 and 528 I0421 10:42:23.182723 47076539613632 model_training_utils.py:505] Train Step: 529/2100 / loss = 1.2474365234375 I0421 10:42:23.183154 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 528 and 529 I0421 10:42:24.268548 47076539613632 model_training_utils.py:505] Train Step: 530/2100 / loss = 1.4296875 I0421 10:42:24.268984 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 529 and 530 I0421 10:42:25.355668 47076539613632 model_training_utils.py:505] Train Step: 531/2100 / loss = 1.753173828125 I0421 10:42:25.356081 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 530 and 531 I0421 10:42:26.438841 47076539613632 model_training_utils.py:505] Train Step: 532/2100 / loss = 2.24072265625 I0421 10:42:26.439264 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 531 and 532 I0421 10:42:27.523907 47076539613632 model_training_utils.py:505] Train Step: 533/2100 / loss = 1.221923828125 I0421 10:42:27.524336 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 532 and 533 I0421 10:42:28.609352 47076539613632 model_training_utils.py:505] Train Step: 534/2100 / loss = 1.296875 I0421 10:42:28.609775 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 533 and 534 I0421 10:42:29.694147 47076539613632 model_training_utils.py:505] Train Step: 535/2100 / loss = 1.3255615234375 I0421 10:42:29.694587 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 534 and 535 I0421 10:42:30.785215 47076539613632 model_training_utils.py:505] Train Step: 536/2100 / loss = 1.50048828125 I0421 10:42:30.785649 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.20 examples/second between steps 535 and 536 I0421 10:42:31.871887 47076539613632 model_training_utils.py:505] Train Step: 537/2100 / loss = 1.531005859375 I0421 10:42:31.872328 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 536 and 537 I0421 10:42:32.957026 47076539613632 model_training_utils.py:505] Train Step: 538/2100 / loss = 1.471923828125 I0421 10:42:32.957467 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 537 and 538 I0421 10:42:34.039138 47076539613632 model_training_utils.py:505] Train Step: 539/2100 / loss = 1.167236328125 I0421 10:42:34.039576 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 538 and 539 I0421 10:42:35.125748 47076539613632 model_training_utils.py:505] Train Step: 540/2100 / loss = 1.1864013671875 I0421 10:42:35.126177 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 539 and 540 I0421 10:42:36.209130 47076539613632 model_training_utils.py:505] Train Step: 541/2100 / loss = 1.158935546875 I0421 10:42:36.209562 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 540 and 541 I0421 10:42:37.294137 47076539613632 model_training_utils.py:505] Train Step: 542/2100 / loss = 1.239501953125 I0421 10:42:37.294570 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 541 and 542 I0421 10:42:38.377949 47076539613632 model_training_utils.py:505] Train Step: 543/2100 / loss = 1.319091796875 I0421 10:42:38.378377 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 542 and 543 I0421 10:42:39.466773 47076539613632 model_training_utils.py:505] Train Step: 544/2100 / loss = 1.2958984375 I0421 10:42:39.467192 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.47 examples/second between steps 543 and 544 I0421 10:42:40.549877 47076539613632 model_training_utils.py:505] Train Step: 545/2100 / loss = 1.6591796875 I0421 10:42:40.550305 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 544 and 545 I0421 10:42:41.640952 47076539613632 model_training_utils.py:505] Train Step: 546/2100 / loss = 1.74462890625 I0421 10:42:41.641383 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.20 examples/second between steps 545 and 546 I0421 10:42:42.721530 47076539613632 model_training_utils.py:505] Train Step: 547/2100 / loss = 1.71875 I0421 10:42:42.721963 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.38 examples/second between steps 546 and 547 I0421 10:42:43.804792 47076539613632 model_training_utils.py:505] Train Step: 548/2100 / loss = 1.993896484375 I0421 10:42:43.805211 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.07 examples/second between steps 547 and 548 I0421 10:42:44.890549 47076539613632 model_training_utils.py:505] Train Step: 549/2100 / loss = 1.3466796875 I0421 10:42:44.890987 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 548 and 549 I0421 10:42:45.973413 47076539613632 model_training_utils.py:505] Train Step: 550/2100 / loss = 1.53955078125 I0421 10:42:45.973851 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 549 and 550 I0421 10:42:47.055010 47076539613632 model_training_utils.py:505] Train Step: 551/2100 / loss = 1.580322265625 I0421 10:42:47.055442 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 550 and 551 I0421 10:42:48.135042 47076539613632 model_training_utils.py:505] Train Step: 552/2100 / loss = 1.4345703125 I0421 10:42:48.135479 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.45 examples/second between steps 551 and 552 I0421 10:42:49.217761 47076539613632 model_training_utils.py:505] Train Step: 553/2100 / loss = 1.1700439453125 I0421 10:42:49.218202 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 552 and 553 I0421 10:42:50.299826 47076539613632 model_training_utils.py:505] Train Step: 554/2100 / loss = 1.68408203125 I0421 10:42:50.300253 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 553 and 554 I0421 10:42:51.381032 47076539613632 model_training_utils.py:505] Train Step: 555/2100 / loss = 1.6318359375 I0421 10:42:51.381459 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.29 examples/second between steps 554 and 555 I0421 10:42:52.465617 47076539613632 model_training_utils.py:505] Train Step: 556/2100 / loss = 1.734375 I0421 10:42:52.466033 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 555 and 556 I0421 10:42:53.549742 47076539613632 model_training_utils.py:505] Train Step: 557/2100 / loss = 1.263671875 I0421 10:42:53.550126 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 556 and 557 I0421 10:42:54.634962 47076539613632 model_training_utils.py:505] Train Step: 558/2100 / loss = 1.4140625 I0421 10:42:54.635354 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 557 and 558 I0421 10:42:55.721018 47076539613632 model_training_utils.py:505] Train Step: 559/2100 / loss = 1.510986328125 I0421 10:42:55.721416 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 558 and 559 I0421 10:42:56.804304 47076539613632 model_training_utils.py:505] Train Step: 560/2100 / loss = 1.947509765625 I0421 10:42:56.804694 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 559 and 560 I0421 10:42:57.890010 47076539613632 model_training_utils.py:505] Train Step: 561/2100 / loss = 2.6337890625 I0421 10:42:57.890402 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 560 and 561 I0421 10:42:58.972122 47076539613632 model_training_utils.py:505] Train Step: 562/2100 / loss = 3.4287109375 I0421 10:42:58.972515 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 561 and 562 I0421 10:43:00.055675 47076539613632 model_training_utils.py:505] Train Step: 563/2100 / loss = 2.85302734375 I0421 10:43:00.056059 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 562 and 563 I0421 10:43:01.142221 47076539613632 model_training_utils.py:505] Train Step: 564/2100 / loss = 2.18310546875 I0421 10:43:01.142615 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 563 and 564 I0421 10:43:02.226521 47076539613632 model_training_utils.py:505] Train Step: 565/2100 / loss = 2.36083984375 I0421 10:43:02.226897 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 564 and 565 I0421 10:43:03.311867 47076539613632 model_training_utils.py:505] Train Step: 566/2100 / loss = 2.0771484375 I0421 10:43:03.312247 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 565 and 566 I0421 10:43:04.395390 47076539613632 model_training_utils.py:505] Train Step: 567/2100 / loss = 1.5810546875 I0421 10:43:04.395780 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 566 and 567 I0421 10:43:05.477968 47076539613632 model_training_utils.py:505] Train Step: 568/2100 / loss = 1.396484375 I0421 10:43:05.478355 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 567 and 568 I0421 10:43:06.563596 47076539613632 model_training_utils.py:505] Train Step: 569/2100 / loss = 1.07080078125 I0421 10:43:06.563980 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 568 and 569 I0421 10:43:07.648089 47076539613632 model_training_utils.py:505] Train Step: 570/2100 / loss = 1.364501953125 I0421 10:43:07.648482 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 569 and 570 I0421 10:43:08.730465 47076539613632 model_training_utils.py:505] Train Step: 571/2100 / loss = 1.3291015625 I0421 10:43:08.730849 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 570 and 571 I0421 10:43:09.814360 47076539613632 model_training_utils.py:505] Train Step: 572/2100 / loss = 1.571533203125 I0421 10:43:09.814746 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 571 and 572 I0421 10:43:10.902310 47076539613632 model_training_utils.py:505] Train Step: 573/2100 / loss = 1.53369140625 I0421 10:43:10.902704 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.55 examples/second between steps 572 and 573 I0421 10:43:11.988309 47076539613632 model_training_utils.py:505] Train Step: 574/2100 / loss = 1.695068359375 I0421 10:43:11.988693 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 573 and 574 I0421 10:43:13.078194 47076539613632 model_training_utils.py:505] Train Step: 575/2100 / loss = 1.4853515625 I0421 10:43:13.078588 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.35 examples/second between steps 574 and 575 I0421 10:43:14.169315 47076539613632 model_training_utils.py:505] Train Step: 576/2100 / loss = 1.5595703125 I0421 10:43:14.169705 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.20 examples/second between steps 575 and 576 I0421 10:43:15.255740 47076539613632 model_training_utils.py:505] Train Step: 577/2100 / loss = 2.1953125 I0421 10:43:15.256119 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 576 and 577 I0421 10:43:16.344471 47076539613632 model_training_utils.py:505] Train Step: 578/2100 / loss = 2.89453125 I0421 10:43:16.344852 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.46 examples/second between steps 577 and 578 I0421 10:43:17.432602 47076539613632 model_training_utils.py:505] Train Step: 579/2100 / loss = 1.6064453125 I0421 10:43:17.432984 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.54 examples/second between steps 578 and 579 I0421 10:43:18.523050 47076539613632 model_training_utils.py:505] Train Step: 580/2100 / loss = 1.244384765625 I0421 10:43:18.523437 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.29 examples/second between steps 579 and 580 I0421 10:43:19.611999 47076539613632 model_training_utils.py:505] Train Step: 581/2100 / loss = 1.341796875 I0421 10:43:19.612398 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.44 examples/second between steps 580 and 581 I0421 10:43:20.699590 47076539613632 model_training_utils.py:505] Train Step: 582/2100 / loss = 1.134033203125 I0421 10:43:20.699985 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.57 examples/second between steps 581 and 582 I0421 10:43:21.787599 47076539613632 model_training_utils.py:505] Train Step: 583/2100 / loss = 1.216064453125 I0421 10:43:21.787975 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.52 examples/second between steps 582 and 583 I0421 10:43:22.875586 47076539613632 model_training_utils.py:505] Train Step: 584/2100 / loss = 1.3050537109375 I0421 10:43:22.875967 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.55 examples/second between steps 583 and 584 I0421 10:43:23.962793 47076539613632 model_training_utils.py:505] Train Step: 585/2100 / loss = 1.6875 I0421 10:43:23.963177 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.62 examples/second between steps 584 and 585 I0421 10:43:25.053447 47076539613632 model_training_utils.py:505] Train Step: 586/2100 / loss = 1.836181640625 I0421 10:43:25.053831 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.25 examples/second between steps 585 and 586 I0421 10:43:26.139171 47076539613632 model_training_utils.py:505] Train Step: 587/2100 / loss = 1.708251953125 I0421 10:43:26.139564 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 586 and 587 I0421 10:43:27.221878 47076539613632 model_training_utils.py:505] Train Step: 588/2100 / loss = 1.346435546875 I0421 10:43:27.222260 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 587 and 588 I0421 10:43:28.310820 47076539613632 model_training_utils.py:505] Train Step: 589/2100 / loss = 1.316162109375 I0421 10:43:28.311199 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.43 examples/second between steps 588 and 589 I0421 10:43:29.395480 47076539613632 model_training_utils.py:505] Train Step: 590/2100 / loss = 2.57275390625 I0421 10:43:29.395864 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 589 and 590 I0421 10:43:30.478429 47076539613632 model_training_utils.py:505] Train Step: 591/2100 / loss = 2.89013671875 I0421 10:43:30.478818 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 590 and 591 I0421 10:43:31.561298 47076539613632 model_training_utils.py:505] Train Step: 592/2100 / loss = 2.180419921875 I0421 10:43:31.561680 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 591 and 592 I0421 10:43:32.644455 47076539613632 model_training_utils.py:505] Train Step: 593/2100 / loss = 1.87060546875 I0421 10:43:32.644843 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 592 and 593 I0421 10:43:33.730708 47076539613632 model_training_utils.py:505] Train Step: 594/2100 / loss = 1.50048828125 I0421 10:43:33.731093 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 593 and 594 I0421 10:43:34.821319 47076539613632 model_training_utils.py:505] Train Step: 595/2100 / loss = 1.3671875 I0421 10:43:34.821706 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.27 examples/second between steps 594 and 595 I0421 10:43:35.908874 47076539613632 model_training_utils.py:505] Train Step: 596/2100 / loss = 1.509521484375 I0421 10:43:35.909252 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 595 and 596 I0421 10:43:36.999645 47076539613632 model_training_utils.py:505] Train Step: 597/2100 / loss = 1.576171875 I0421 10:43:37.000028 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.26 examples/second between steps 596 and 597 I0421 10:43:38.087806 47076539613632 model_training_utils.py:505] Train Step: 598/2100 / loss = 1.548583984375 I0421 10:43:38.088185 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.51 examples/second between steps 597 and 598 I0421 10:43:39.171206 47076539613632 model_training_utils.py:505] Train Step: 599/2100 / loss = 1.49169921875 I0421 10:43:39.171595 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 598 and 599 I0421 10:43:40.257179 47076539613632 model_training_utils.py:505] Train Step: 600/2100 / loss = 1.656494140625 I0421 10:43:40.257563 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 599 and 600 I0421 10:43:41.342077 47076539613632 model_training_utils.py:505] Train Step: 601/2100 / loss = 1.6357421875 I0421 10:43:41.342473 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 600 and 601 I0421 10:43:42.429971 47076539613632 model_training_utils.py:505] Train Step: 602/2100 / loss = 2.330078125 I0421 10:43:42.430367 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.54 examples/second between steps 601 and 602 I0421 10:43:43.520302 47076539613632 model_training_utils.py:505] Train Step: 603/2100 / loss = 1.98681640625 I0421 10:43:43.520684 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.28 examples/second between steps 602 and 603 I0421 10:43:44.603439 47076539613632 model_training_utils.py:505] Train Step: 604/2100 / loss = 1.5146484375 I0421 10:43:44.603826 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.07 examples/second between steps 603 and 604 I0421 10:43:45.681530 47076539613632 model_training_utils.py:505] Train Step: 605/2100 / loss = 1.34765625 I0421 10:43:45.681915 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.65 examples/second between steps 604 and 605 I0421 10:43:46.764645 47076539613632 model_training_utils.py:505] Train Step: 606/2100 / loss = 1.315673828125 I0421 10:43:46.765041 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 605 and 606 I0421 10:43:47.848131 47076539613632 model_training_utils.py:505] Train Step: 607/2100 / loss = 1.25390625 I0421 10:43:47.848522 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 606 and 607 I0421 10:43:48.932395 47076539613632 model_training_utils.py:505] Train Step: 608/2100 / loss = 1.201416015625 I0421 10:43:48.932774 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 607 and 608 I0421 10:43:50.018388 47076539613632 model_training_utils.py:505] Train Step: 609/2100 / loss = 1.0877685546875 I0421 10:43:50.018774 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 608 and 609 I0421 10:43:51.105645 47076539613632 model_training_utils.py:505] Train Step: 610/2100 / loss = 1.391845703125 I0421 10:43:51.106032 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 609 and 610 I0421 10:43:52.193206 47076539613632 model_training_utils.py:505] Train Step: 611/2100 / loss = 1.63671875 I0421 10:43:52.193593 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 610 and 611 I0421 10:43:53.279977 47076539613632 model_training_utils.py:505] Train Step: 612/2100 / loss = 1.53271484375 I0421 10:43:53.280368 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 611 and 612 I0421 10:43:54.366604 47076539613632 model_training_utils.py:505] Train Step: 613/2100 / loss = 1.13818359375 I0421 10:43:54.367000 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 612 and 613 I0421 10:43:55.448639 47076539613632 model_training_utils.py:505] Train Step: 614/2100 / loss = 1.273681640625 I0421 10:43:55.449026 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 613 and 614 I0421 10:43:56.533185 47076539613632 model_training_utils.py:505] Train Step: 615/2100 / loss = 1.1884765625 I0421 10:43:56.533572 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 614 and 615 I0421 10:43:57.621979 47076539613632 model_training_utils.py:505] Train Step: 616/2100 / loss = 1.409912109375 I0421 10:43:57.622370 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.51 examples/second between steps 615 and 616 I0421 10:43:58.706182 47076539613632 model_training_utils.py:505] Train Step: 617/2100 / loss = 1.811767578125 I0421 10:43:58.706586 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 616 and 617 I0421 10:43:59.793736 47076539613632 model_training_utils.py:505] Train Step: 618/2100 / loss = 1.96826171875 I0421 10:43:59.794120 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.58 examples/second between steps 617 and 618 I0421 10:44:00.874525 47076539613632 model_training_utils.py:505] Train Step: 619/2100 / loss = 1.72900390625 I0421 10:44:00.874905 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.31 examples/second between steps 618 and 619 I0421 10:44:01.966198 47076539613632 model_training_utils.py:505] Train Step: 620/2100 / loss = 2.293212890625 I0421 10:44:01.966587 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.11 examples/second between steps 619 and 620 I0421 10:44:03.048389 47076539613632 model_training_utils.py:505] Train Step: 621/2100 / loss = 2.88134765625 I0421 10:44:03.048772 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 620 and 621 I0421 10:44:04.134586 47076539613632 model_training_utils.py:505] Train Step: 622/2100 / loss = 2.81396484375 I0421 10:44:04.134976 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 621 and 622 I0421 10:44:05.217406 47076539613632 model_training_utils.py:505] Train Step: 623/2100 / loss = 1.847900390625 I0421 10:44:05.217785 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 622 and 623 I0421 10:44:06.300464 47076539613632 model_training_utils.py:505] Train Step: 624/2100 / loss = 1.392578125 I0421 10:44:06.300846 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 623 and 624 I0421 10:44:07.384650 47076539613632 model_training_utils.py:505] Train Step: 625/2100 / loss = 1.62060546875 I0421 10:44:07.385029 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 624 and 625 I0421 10:44:08.466833 47076539613632 model_training_utils.py:505] Train Step: 626/2100 / loss = 1.36376953125 I0421 10:44:08.467219 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 625 and 626 I0421 10:44:09.550914 47076539613632 model_training_utils.py:505] Train Step: 627/2100 / loss = 1.07861328125 I0421 10:44:09.551310 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 626 and 627 I0421 10:44:10.633571 47076539613632 model_training_utils.py:505] Train Step: 628/2100 / loss = 1.080322265625 I0421 10:44:10.633949 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 627 and 628 I0421 10:44:11.717386 47076539613632 model_training_utils.py:505] Train Step: 629/2100 / loss = 1.184326171875 I0421 10:44:11.717762 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 628 and 629 I0421 10:44:12.804082 47076539613632 model_training_utils.py:505] Train Step: 630/2100 / loss = 1.21533203125 I0421 10:44:12.804469 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 629 and 630 I0421 10:44:13.889343 47076539613632 model_training_utils.py:505] Train Step: 631/2100 / loss = 1.33203125 I0421 10:44:13.889723 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 630 and 631 I0421 10:44:14.971918 47076539613632 model_training_utils.py:505] Train Step: 632/2100 / loss = 1.25927734375 I0421 10:44:14.972309 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 631 and 632 I0421 10:44:16.054768 47076539613632 model_training_utils.py:505] Train Step: 633/2100 / loss = 1.188720703125 I0421 10:44:16.055152 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 632 and 633 I0421 10:44:17.138791 47076539613632 model_training_utils.py:505] Train Step: 634/2100 / loss = 1.200927734375 I0421 10:44:17.139173 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 633 and 634 I0421 10:44:18.226878 47076539613632 model_training_utils.py:505] Train Step: 635/2100 / loss = 1.4296875 I0421 10:44:18.227264 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.48 examples/second between steps 634 and 635 I0421 10:44:19.312072 47076539613632 model_training_utils.py:505] Train Step: 636/2100 / loss = 1.25927734375 I0421 10:44:19.312462 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 635 and 636 I0421 10:44:20.391440 47076539613632 model_training_utils.py:505] Train Step: 637/2100 / loss = 1.5615234375 I0421 10:44:20.391822 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.46 examples/second between steps 636 and 637 I0421 10:44:21.477019 47076539613632 model_training_utils.py:505] Train Step: 638/2100 / loss = 1.2210693359375 I0421 10:44:21.477405 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 637 and 638 I0421 10:44:22.559295 47076539613632 model_training_utils.py:505] Train Step: 639/2100 / loss = 1.01220703125 I0421 10:44:22.559678 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 638 and 639 I0421 10:44:23.652218 47076539613632 model_training_utils.py:505] Train Step: 640/2100 / loss = 1.3466796875 I0421 10:44:23.652603 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.01 examples/second between steps 639 and 640 I0421 10:44:24.745176 47076539613632 model_training_utils.py:505] Train Step: 641/2100 / loss = 1.517578125 I0421 10:44:24.745562 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.00 examples/second between steps 640 and 641 I0421 10:44:25.833340 47076539613632 model_training_utils.py:505] Train Step: 642/2100 / loss = 1.434814453125 I0421 10:44:25.833722 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.50 examples/second between steps 641 and 642 I0421 10:44:26.923656 47076539613632 model_training_utils.py:505] Train Step: 643/2100 / loss = 1.504150390625 I0421 10:44:26.924043 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.28 examples/second between steps 642 and 643 I0421 10:44:28.011352 47076539613632 model_training_utils.py:505] Train Step: 644/2100 / loss = 0.82373046875 I0421 10:44:28.011733 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.56 examples/second between steps 643 and 644 I0421 10:44:29.097480 47076539613632 model_training_utils.py:505] Train Step: 645/2100 / loss = 1.203125 I0421 10:44:29.097867 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 644 and 645 I0421 10:44:30.184471 47076539613632 model_training_utils.py:505] Train Step: 646/2100 / loss = 1.51025390625 I0421 10:44:30.184864 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.64 examples/second between steps 645 and 646 I0421 10:44:31.269601 47076539613632 model_training_utils.py:505] Train Step: 647/2100 / loss = 1.08544921875 I0421 10:44:31.269987 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 646 and 647 I0421 10:44:32.350884 47076539613632 model_training_utils.py:505] Train Step: 648/2100 / loss = 1.123779296875 I0421 10:44:32.351275 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 647 and 648 I0421 10:44:33.438771 47076539613632 model_training_utils.py:505] Train Step: 649/2100 / loss = 1.60498046875 I0421 10:44:33.439157 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.62 examples/second between steps 648 and 649 I0421 10:44:34.526220 47076539613632 model_training_utils.py:505] Train Step: 650/2100 / loss = 1.82421875 I0421 10:44:34.526609 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.61 examples/second between steps 649 and 650 I0421 10:44:35.612705 47076539613632 model_training_utils.py:505] Train Step: 651/2100 / loss = 2.574951171875 I0421 10:44:35.613089 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 650 and 651 I0421 10:44:36.703492 47076539613632 model_training_utils.py:505] Train Step: 652/2100 / loss = 1.600341796875 I0421 10:44:36.703873 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.23 examples/second between steps 651 and 652 I0421 10:44:37.789536 47076539613632 model_training_utils.py:505] Train Step: 653/2100 / loss = 1.4482421875 I0421 10:44:37.789913 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 652 and 653 I0421 10:44:38.878498 47076539613632 model_training_utils.py:505] Train Step: 654/2100 / loss = 1.532470703125 I0421 10:44:38.878877 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.41 examples/second between steps 653 and 654 I0421 10:44:39.967799 47076539613632 model_training_utils.py:505] Train Step: 655/2100 / loss = 1.4287109375 I0421 10:44:39.968179 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.41 examples/second between steps 654 and 655 I0421 10:44:41.054824 47076539613632 model_training_utils.py:505] Train Step: 656/2100 / loss = 1.3525390625 I0421 10:44:41.055204 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.62 examples/second between steps 655 and 656 I0421 10:44:42.139502 47076539613632 model_training_utils.py:505] Train Step: 657/2100 / loss = 1.405029296875 I0421 10:44:42.139877 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 656 and 657 I0421 10:44:43.224071 47076539613632 model_training_utils.py:505] Train Step: 658/2100 / loss = 1.1180419921875 I0421 10:44:43.224455 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 657 and 658 I0421 10:44:44.306572 47076539613632 model_training_utils.py:505] Train Step: 659/2100 / loss = 1.318603515625 I0421 10:44:44.306953 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 658 and 659 I0421 10:44:45.387191 47076539613632 model_training_utils.py:505] Train Step: 660/2100 / loss = 1.44384765625 I0421 10:44:45.387579 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.34 examples/second between steps 659 and 660 I0421 10:44:46.464014 47076539613632 model_training_utils.py:505] Train Step: 661/2100 / loss = 1.50732421875 I0421 10:44:46.464403 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.78 examples/second between steps 660 and 661 I0421 10:44:47.546672 47076539613632 model_training_utils.py:505] Train Step: 662/2100 / loss = 1.35498046875 I0421 10:44:47.547060 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 661 and 662 I0421 10:44:48.631193 47076539613632 model_training_utils.py:505] Train Step: 663/2100 / loss = 1.14892578125 I0421 10:44:48.631589 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 662 and 663 I0421 10:44:49.717002 47076539613632 model_training_utils.py:505] Train Step: 664/2100 / loss = 0.997314453125 I0421 10:44:49.717395 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 663 and 664 I0421 10:44:50.804101 47076539613632 model_training_utils.py:505] Train Step: 665/2100 / loss = 1.22509765625 I0421 10:44:50.804491 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.62 examples/second between steps 664 and 665 I0421 10:44:51.894896 47076539613632 model_training_utils.py:505] Train Step: 666/2100 / loss = 1.222412109375 I0421 10:44:51.895291 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.23 examples/second between steps 665 and 666 I0421 10:44:52.979215 47076539613632 model_training_utils.py:505] Train Step: 667/2100 / loss = 1.14501953125 I0421 10:44:52.979614 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 666 and 667 I0421 10:44:54.065119 47076539613632 model_training_utils.py:505] Train Step: 668/2100 / loss = 1.451416015625 I0421 10:44:54.065507 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 667 and 668 I0421 10:44:55.151186 47076539613632 model_training_utils.py:505] Train Step: 669/2100 / loss = 1.565185546875 I0421 10:44:55.151571 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 668 and 669 I0421 10:44:56.235175 47076539613632 model_training_utils.py:505] Train Step: 670/2100 / loss = 1.585693359375 I0421 10:44:56.235565 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 669 and 670 I0421 10:44:57.322276 47076539613632 model_training_utils.py:505] Train Step: 671/2100 / loss = 1.7314453125 I0421 10:44:57.322689 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.63 examples/second between steps 670 and 671 I0421 10:44:58.405473 47076539613632 model_training_utils.py:505] Train Step: 672/2100 / loss = 1.696044921875 I0421 10:44:58.405854 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 671 and 672 I0421 10:44:59.492201 47076539613632 model_training_utils.py:505] Train Step: 673/2100 / loss = 1.277099609375 I0421 10:44:59.492589 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.70 examples/second between steps 672 and 673 I0421 10:45:00.579190 47076539613632 model_training_utils.py:505] Train Step: 674/2100 / loss = 1.24853515625 I0421 10:45:00.579579 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 673 and 674 I0421 10:45:01.670888 47076539613632 model_training_utils.py:505] Train Step: 675/2100 / loss = 1.3271484375 I0421 10:45:01.671272 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.14 examples/second between steps 674 and 675 I0421 10:45:02.757252 47076539613632 model_training_utils.py:505] Train Step: 676/2100 / loss = 0.95849609375 I0421 10:45:02.757650 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 675 and 676 I0421 10:45:03.845877 47076539613632 model_training_utils.py:505] Train Step: 677/2100 / loss = 1.044677734375 I0421 10:45:03.846261 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.48 examples/second between steps 676 and 677 I0421 10:45:04.922216 47076539613632 model_training_utils.py:505] Train Step: 678/2100 / loss = 1.209228515625 I0421 10:45:04.922612 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.81 examples/second between steps 677 and 678 I0421 10:45:06.008221 47076539613632 model_training_utils.py:505] Train Step: 679/2100 / loss = 1.0667724609375 I0421 10:45:06.008608 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 678 and 679 I0421 10:45:07.094058 47076539613632 model_training_utils.py:505] Train Step: 680/2100 / loss = 1.12890625 I0421 10:45:07.094449 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 679 and 680 I0421 10:45:08.180089 47076539613632 model_training_utils.py:505] Train Step: 681/2100 / loss = 1.7080078125 I0421 10:45:08.180488 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 680 and 681 I0421 10:45:09.262519 47076539613632 model_training_utils.py:505] Train Step: 682/2100 / loss = 1.328857421875 I0421 10:45:09.262900 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 681 and 682 I0421 10:45:10.343711 47076539613632 model_training_utils.py:505] Train Step: 683/2100 / loss = 1.04931640625 I0421 10:45:10.344092 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.31 examples/second between steps 682 and 683 I0421 10:45:11.428914 47076539613632 model_training_utils.py:505] Train Step: 684/2100 / loss = 1.166748046875 I0421 10:45:11.429310 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 683 and 684 I0421 10:45:12.516049 47076539613632 model_training_utils.py:505] Train Step: 685/2100 / loss = 1.127685546875 I0421 10:45:12.516457 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 684 and 685 I0421 10:45:13.600478 47076539613632 model_training_utils.py:505] Train Step: 686/2100 / loss = 1.184326171875 I0421 10:45:13.600865 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 685 and 686 I0421 10:45:14.685941 47076539613632 model_training_utils.py:505] Train Step: 687/2100 / loss = 1.1776123046875 I0421 10:45:14.686342 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 686 and 687 I0421 10:45:15.774722 47076539613632 model_training_utils.py:505] Train Step: 688/2100 / loss = 1.32373046875 I0421 10:45:15.775116 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.43 examples/second between steps 687 and 688 I0421 10:45:16.859976 47076539613632 model_training_utils.py:505] Train Step: 689/2100 / loss = 1.2239990234375 I0421 10:45:16.860368 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 688 and 689 I0421 10:45:17.941788 47076539613632 model_training_utils.py:505] Train Step: 690/2100 / loss = 1.71875 I0421 10:45:17.942172 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 689 and 690 I0421 10:45:19.024093 47076539613632 model_training_utils.py:505] Train Step: 691/2100 / loss = 1.412109375 I0421 10:45:19.024484 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 690 and 691 I0421 10:45:20.109707 47076539613632 model_training_utils.py:505] Train Step: 692/2100 / loss = 1.465087890625 I0421 10:45:20.110089 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 691 and 692 I0421 10:45:21.198521 47076539613632 model_training_utils.py:505] Train Step: 693/2100 / loss = 1.39208984375 I0421 10:45:21.198901 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.43 examples/second between steps 692 and 693 I0421 10:45:22.284507 47076539613632 model_training_utils.py:505] Train Step: 694/2100 / loss = 1.365478515625 I0421 10:45:22.284894 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 693 and 694 I0421 10:45:23.369710 47076539613632 model_training_utils.py:505] Train Step: 695/2100 / loss = 1.259765625 I0421 10:45:23.370094 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 694 and 695 I0421 10:45:24.455134 47076539613632 model_training_utils.py:505] Train Step: 696/2100 / loss = 1.013916015625 I0421 10:45:24.455540 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 695 and 696 I0421 10:45:25.540038 47076539613632 model_training_utils.py:505] Train Step: 697/2100 / loss = 0.97607421875 I0421 10:45:25.540440 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 696 and 697 I0421 10:45:26.621473 47076539613632 model_training_utils.py:505] Train Step: 698/2100 / loss = 1.000244140625 I0421 10:45:26.621858 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.25 examples/second between steps 697 and 698 I0421 10:45:27.703630 47076539613632 model_training_utils.py:505] Train Step: 699/2100 / loss = 1.166748046875 I0421 10:45:27.704011 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 698 and 699 I0421 10:45:28.789150 47076539613632 model_training_utils.py:505] Train Step: 700/2100 / loss = 1.030029296875 I0421 10:45:37.176540 47076539613632 model_training_utils.py:49] Saving model as TF checkpoint: /public/home/xuanbaby/DL-TensorFlow/models_r2.3.0/official/nlp/bert/model_squad_v2/ctl_step_700.ckpt-1 I0421 10:45:37.408300 47076539613632 keras_utils.py:133] TimeHistory: 9.58 seconds, 13.36 examples/second between steps 699 and 700 I0421 10:45:38.737152 47076539613632 model_training_utils.py:505] Train Step: 701/2100 / loss = 1.2469482421875 I0421 10:45:38.738032 47076539613632 keras_utils.py:133] TimeHistory: 1.21 seconds, 74354.83 examples/second between steps 700 and 1401 I0421 10:45:39.820216 47076539613632 model_training_utils.py:505] Train Step: 702/2100 / loss = 0.9532470703125 I0421 10:45:39.820623 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 1401 and 1402 I0421 10:45:40.902045 47076539613632 model_training_utils.py:505] Train Step: 703/2100 / loss = 1.380126953125 I0421 10:45:40.902448 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 1402 and 1403 I0421 10:45:41.989025 47076539613632 model_training_utils.py:505] Train Step: 704/2100 / loss = 1.504638671875 I0421 10:45:41.989432 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.67 examples/second between steps 1403 and 1404 I0421 10:45:43.075928 47076539613632 model_training_utils.py:505] Train Step: 705/2100 / loss = 1.27392578125 I0421 10:45:43.076340 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 1404 and 1405 I0421 10:45:44.162086 47076539613632 model_training_utils.py:505] Train Step: 706/2100 / loss = 1.550048828125 I0421 10:45:44.162497 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 1405 and 1406 I0421 10:45:45.247252 47076539613632 model_training_utils.py:505] Train Step: 707/2100 / loss = 1.4892578125 I0421 10:45:45.247696 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 1406 and 1407 I0421 10:45:46.335277 47076539613632 model_training_utils.py:505] Train Step: 708/2100 / loss = 1.344482421875 I0421 10:45:46.335704 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 1407 and 1408 I0421 10:45:47.421876 47076539613632 model_training_utils.py:505] Train Step: 709/2100 / loss = 1.0712890625 I0421 10:45:47.422302 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 1408 and 1409 I0421 10:45:48.509137 47076539613632 model_training_utils.py:505] Train Step: 710/2100 / loss = 1.25048828125 I0421 10:45:48.509560 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.63 examples/second between steps 1409 and 1410 I0421 10:45:49.596687 47076539613632 model_training_utils.py:505] Train Step: 711/2100 / loss = 1.4609375 I0421 10:45:49.597111 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.61 examples/second between steps 1410 and 1411 I0421 10:45:50.679113 47076539613632 model_training_utils.py:505] Train Step: 712/2100 / loss = 1.628662109375 I0421 10:45:50.679545 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 1411 and 1412 I0421 10:45:51.764613 47076539613632 model_training_utils.py:505] Train Step: 713/2100 / loss = 1.073486328125 I0421 10:45:51.765036 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 1412 and 1413 I0421 10:45:52.849865 47076539613632 model_training_utils.py:505] Train Step: 714/2100 / loss = 1.19140625 I0421 10:45:52.850298 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 1413 and 1414 I0421 10:45:53.935842 47076539613632 model_training_utils.py:505] Train Step: 715/2100 / loss = 1.3719482421875 I0421 10:45:53.936269 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 1414 and 1415 I0421 10:45:55.023673 47076539613632 model_training_utils.py:505] Train Step: 716/2100 / loss = 1.1033935546875 I0421 10:45:55.024100 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.54 examples/second between steps 1415 and 1416 I0421 10:45:56.112215 47076539613632 model_training_utils.py:505] Train Step: 717/2100 / loss = 1.0306396484375 I0421 10:45:56.112655 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.48 examples/second between steps 1416 and 1417 I0421 10:45:57.197658 47076539613632 model_training_utils.py:505] Train Step: 718/2100 / loss = 1.209716796875 I0421 10:45:57.198084 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 1417 and 1418 I0421 10:45:58.280993 47076539613632 model_training_utils.py:505] Train Step: 719/2100 / loss = 1.626220703125 I0421 10:45:58.281429 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 1418 and 1419 I0421 10:45:59.364000 47076539613632 model_training_utils.py:505] Train Step: 720/2100 / loss = 1.612548828125 I0421 10:45:59.364407 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 1419 and 1420 I0421 10:46:00.462738 47076539613632 model_training_utils.py:505] Train Step: 721/2100 / loss = 1.56884765625 I0421 10:46:00.463166 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.32 examples/second between steps 1420 and 1421 I0421 10:46:01.556326 47076539613632 model_training_utils.py:505] Train Step: 722/2100 / loss = 1.491943359375 I0421 10:46:01.556756 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 117.98 examples/second between steps 1421 and 1422 I0421 10:46:02.648989 47076539613632 model_training_utils.py:505] Train Step: 723/2100 / loss = 1.358154296875 I0421 10:46:02.649419 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.06 examples/second between steps 1422 and 1423 I0421 10:46:03.731360 47076539613632 model_training_utils.py:505] Train Step: 724/2100 / loss = 1.244140625 I0421 10:46:03.731771 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 1423 and 1424 I0421 10:46:04.820967 47076539613632 model_training_utils.py:505] Train Step: 725/2100 / loss = 1.341552734375 I0421 10:46:04.821419 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.37 examples/second between steps 1424 and 1425 I0421 10:46:05.908252 47076539613632 model_training_utils.py:505] Train Step: 726/2100 / loss = 1.3779296875 I0421 10:46:05.908677 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 1425 and 1426 I0421 10:46:06.991475 47076539613632 model_training_utils.py:505] Train Step: 727/2100 / loss = 1.47314453125 I0421 10:46:06.991911 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 1426 and 1427 I0421 10:46:08.084931 47076539613632 model_training_utils.py:505] Train Step: 728/2100 / loss = 1.179443359375 I0421 10:46:08.085366 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.92 examples/second between steps 1427 and 1428 I0421 10:46:09.180355 47076539613632 model_training_utils.py:505] Train Step: 729/2100 / loss = 1.417724609375 I0421 10:46:09.180784 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.74 examples/second between steps 1428 and 1429 I0421 10:46:10.269257 47076539613632 model_training_utils.py:505] Train Step: 730/2100 / loss = 1.31298828125 I0421 10:46:10.269693 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.44 examples/second between steps 1429 and 1430 I0421 10:46:11.355356 47076539613632 model_training_utils.py:505] Train Step: 731/2100 / loss = 1.1357421875 I0421 10:46:11.355787 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 1430 and 1431 I0421 10:46:12.447300 47076539613632 model_training_utils.py:505] Train Step: 732/2100 / loss = 0.833251953125 I0421 10:46:12.447731 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.09 examples/second between steps 1431 and 1432 I0421 10:46:13.539188 47076539613632 model_training_utils.py:505] Train Step: 733/2100 / loss = 1.05419921875 I0421 10:46:13.539617 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.11 examples/second between steps 1432 and 1433 I0421 10:46:14.627042 47076539613632 model_training_utils.py:505] Train Step: 734/2100 / loss = 1.1300048828125 I0421 10:46:14.627472 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.64 examples/second between steps 1433 and 1434 I0421 10:46:15.715484 47076539613632 model_training_utils.py:505] Train Step: 735/2100 / loss = 0.992919921875 I0421 10:46:15.715905 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.51 examples/second between steps 1434 and 1435 I0421 10:46:16.807507 47076539613632 model_training_utils.py:505] Train Step: 736/2100 / loss = 1.007568359375 I0421 10:46:16.807928 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.13 examples/second between steps 1435 and 1436 I0421 10:46:17.898707 47076539613632 model_training_utils.py:505] Train Step: 737/2100 / loss = 0.8719482421875 I0421 10:46:17.899128 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.21 examples/second between steps 1436 and 1437 I0421 10:46:18.986259 47076539613632 model_training_utils.py:505] Train Step: 738/2100 / loss = 1.1715087890625 I0421 10:46:18.986695 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.59 examples/second between steps 1437 and 1438 I0421 10:46:20.072453 47076539613632 model_training_utils.py:505] Train Step: 739/2100 / loss = 1.297607421875 I0421 10:46:20.072876 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 1438 and 1439 I0421 10:46:21.160135 47076539613632 model_training_utils.py:505] Train Step: 740/2100 / loss = 1.576904296875 I0421 10:46:21.160567 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.55 examples/second between steps 1439 and 1440 I0421 10:46:22.249179 47076539613632 model_training_utils.py:505] Train Step: 741/2100 / loss = 1.447998046875 I0421 10:46:22.249614 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.43 examples/second between steps 1440 and 1441 I0421 10:46:23.331606 47076539613632 model_training_utils.py:505] Train Step: 742/2100 / loss = 1.65771484375 I0421 10:46:23.332027 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 1441 and 1442 I0421 10:46:24.417267 47076539613632 model_training_utils.py:505] Train Step: 743/2100 / loss = 1.41455078125 I0421 10:46:24.417701 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 1442 and 1443 I0421 10:46:25.506669 47076539613632 model_training_utils.py:505] Train Step: 744/2100 / loss = 1.18115234375 I0421 10:46:25.507083 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.37 examples/second between steps 1443 and 1444 I0421 10:46:26.592749 47076539613632 model_training_utils.py:505] Train Step: 745/2100 / loss = 1.316650390625 I0421 10:46:26.593169 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 1444 and 1445 I0421 10:46:27.679553 47076539613632 model_training_utils.py:505] Train Step: 746/2100 / loss = 1.0540771484375 I0421 10:46:27.679975 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.67 examples/second between steps 1445 and 1446 I0421 10:46:28.768438 47076539613632 model_training_utils.py:505] Train Step: 747/2100 / loss = 1.141357421875 I0421 10:46:28.768857 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.45 examples/second between steps 1446 and 1447 I0421 10:46:29.855354 47076539613632 model_training_utils.py:505] Train Step: 748/2100 / loss = 1.465576171875 I0421 10:46:29.855776 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 1447 and 1448 I0421 10:46:30.939173 47076539613632 model_training_utils.py:505] Train Step: 749/2100 / loss = 1.004638671875 I0421 10:46:30.939602 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 1448 and 1449 I0421 10:46:32.023627 47076539613632 model_training_utils.py:505] Train Step: 750/2100 / loss = 1.531494140625 I0421 10:46:32.024043 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 1449 and 1450 I0421 10:46:33.108814 47076539613632 model_training_utils.py:505] Train Step: 751/2100 / loss = 1.543212890625 I0421 10:46:33.109235 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 1450 and 1451 I0421 10:46:34.200939 47076539613632 model_training_utils.py:505] Train Step: 752/2100 / loss = 1.673828125 I0421 10:46:34.201373 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.14 examples/second between steps 1451 and 1452 I0421 10:46:35.291742 47076539613632 model_training_utils.py:505] Train Step: 753/2100 / loss = 1.551513671875 I0421 10:46:35.292165 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.23 examples/second between steps 1452 and 1453 I0421 10:46:36.382927 47076539613632 model_training_utils.py:505] Train Step: 754/2100 / loss = 1.57275390625 I0421 10:46:36.383347 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.22 examples/second between steps 1453 and 1454 I0421 10:46:37.472649 47076539613632 model_training_utils.py:505] Train Step: 755/2100 / loss = 1.445556640625 I0421 10:46:37.473055 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.34 examples/second between steps 1454 and 1455 I0421 10:46:38.567592 47076539613632 model_training_utils.py:505] Train Step: 756/2100 / loss = 1.4521484375 I0421 10:46:38.568008 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.78 examples/second between steps 1455 and 1456 I0421 10:46:39.649745 47076539613632 model_training_utils.py:505] Train Step: 757/2100 / loss = 1.593994140625 I0421 10:46:39.650160 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 1456 and 1457 I0421 10:46:40.748856 47076539613632 model_training_utils.py:505] Train Step: 758/2100 / loss = 1.371337890625 I0421 10:46:40.749279 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.31 examples/second between steps 1457 and 1458 I0421 10:46:41.846483 47076539613632 model_training_utils.py:505] Train Step: 759/2100 / loss = 1.12548828125 I0421 10:46:41.846907 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.47 examples/second between steps 1458 and 1459 I0421 10:46:42.949684 47076539613632 model_training_utils.py:505] Train Step: 760/2100 / loss = 0.8424072265625 I0421 10:46:42.950146 47076539613632 keras_utils.py:133] TimeHistory: 1.10 seconds, 116.86 examples/second between steps 1459 and 1460 I0421 10:46:44.046392 47076539613632 model_training_utils.py:505] Train Step: 761/2100 / loss = 1.159423828125 I0421 10:46:44.046830 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.62 examples/second between steps 1460 and 1461 I0421 10:46:45.134300 47076539613632 model_training_utils.py:505] Train Step: 762/2100 / loss = 1.418212890625 I0421 10:46:45.134723 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.56 examples/second between steps 1461 and 1462 I0421 10:46:46.233692 47076539613632 model_training_utils.py:505] Train Step: 763/2100 / loss = 1.49365234375 I0421 10:46:46.234076 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.42 examples/second between steps 1462 and 1463 I0421 10:46:47.326311 47076539613632 model_training_utils.py:505] Train Step: 764/2100 / loss = 1.61865234375 I0421 10:46:47.326707 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.04 examples/second between steps 1463 and 1464 I0421 10:46:48.424350 47076539613632 model_training_utils.py:505] Train Step: 765/2100 / loss = 1.299072265625 I0421 10:46:48.424742 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.49 examples/second between steps 1464 and 1465 I0421 10:46:49.514236 47076539613632 model_training_utils.py:505] Train Step: 766/2100 / loss = 1.4912109375 I0421 10:46:49.514643 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.37 examples/second between steps 1465 and 1466 I0421 10:46:50.598412 47076539613632 model_training_utils.py:505] Train Step: 767/2100 / loss = 1.257080078125 I0421 10:46:50.598797 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 1466 and 1467 I0421 10:46:51.680390 47076539613632 model_training_utils.py:505] Train Step: 768/2100 / loss = 1.225830078125 I0421 10:46:51.680774 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 1467 and 1468 I0421 10:46:52.763446 47076539613632 model_training_utils.py:505] Train Step: 769/2100 / loss = 1.21337890625 I0421 10:46:52.763819 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 1468 and 1469 I0421 10:46:53.838383 47076539613632 model_training_utils.py:505] Train Step: 770/2100 / loss = 1.94140625 I0421 10:46:53.838765 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 120.01 examples/second between steps 1469 and 1470 I0421 10:46:54.924060 47076539613632 model_training_utils.py:505] Train Step: 771/2100 / loss = 1.764404296875 I0421 10:46:54.924459 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 1470 and 1471 I0421 10:46:56.009514 47076539613632 model_training_utils.py:505] Train Step: 772/2100 / loss = 2.001220703125 I0421 10:46:56.009897 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 1471 and 1472 I0421 10:46:57.097688 47076539613632 model_training_utils.py:505] Train Step: 773/2100 / loss = 1.507080078125 I0421 10:46:57.098078 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.53 examples/second between steps 1472 and 1473 I0421 10:46:58.182351 47076539613632 model_training_utils.py:505] Train Step: 774/2100 / loss = 0.987060546875 I0421 10:46:58.182735 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 1473 and 1474 I0421 10:46:59.269327 47076539613632 model_training_utils.py:505] Train Step: 775/2100 / loss = 0.9698486328125 I0421 10:46:59.269714 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 1474 and 1475 I0421 10:47:00.363641 47076539613632 model_training_utils.py:505] Train Step: 776/2100 / loss = 1.056396484375 I0421 10:47:00.364024 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.84 examples/second between steps 1475 and 1476 I0421 10:47:01.459914 47076539613632 model_training_utils.py:505] Train Step: 777/2100 / loss = 1.16357421875 I0421 10:47:01.460308 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.63 examples/second between steps 1476 and 1477 I0421 10:47:02.564208 47076539613632 model_training_utils.py:505] Train Step: 778/2100 / loss = 1.0927734375 I0421 10:47:02.564594 47076539613632 keras_utils.py:133] TimeHistory: 1.10 seconds, 116.80 examples/second between steps 1477 and 1478 I0421 10:47:03.656044 47076539613632 model_training_utils.py:505] Train Step: 779/2100 / loss = 1.0516357421875 I0421 10:47:03.656439 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.15 examples/second between steps 1478 and 1479 I0421 10:47:04.755882 47076539613632 model_training_utils.py:505] Train Step: 780/2100 / loss = 0.943359375 I0421 10:47:04.756278 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.25 examples/second between steps 1479 and 1480 I0421 10:47:05.854066 47076539613632 model_training_utils.py:505] Train Step: 781/2100 / loss = 1.196044921875 I0421 10:47:05.854457 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.48 examples/second between steps 1480 and 1481 I0421 10:47:06.953373 47076539613632 model_training_utils.py:505] Train Step: 782/2100 / loss = 1.45947265625 I0421 10:47:06.953804 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.40 examples/second between steps 1481 and 1482 I0421 10:47:08.049830 47076539613632 model_training_utils.py:505] Train Step: 783/2100 / loss = 1.455322265625 I0421 10:47:08.050253 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.64 examples/second between steps 1482 and 1483 I0421 10:47:09.140326 47076539613632 model_training_utils.py:505] Train Step: 784/2100 / loss = 1.352783203125 I0421 10:47:09.140758 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.26 examples/second between steps 1483 and 1484 I0421 10:47:10.230097 47076539613632 model_training_utils.py:505] Train Step: 785/2100 / loss = 1.33544921875 I0421 10:47:10.230528 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.36 examples/second between steps 1484 and 1485 I0421 10:47:11.318606 47076539613632 model_training_utils.py:505] Train Step: 786/2100 / loss = 1.515625 I0421 10:47:11.319035 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.50 examples/second between steps 1485 and 1486 I0421 10:47:12.408524 47076539613632 model_training_utils.py:505] Train Step: 787/2100 / loss = 1.65283203125 I0421 10:47:12.408941 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.35 examples/second between steps 1486 and 1487 I0421 10:47:13.501561 47076539613632 model_training_utils.py:505] Train Step: 788/2100 / loss = 1.241943359375 I0421 10:47:13.501983 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.00 examples/second between steps 1487 and 1488 I0421 10:47:14.607057 47076539613632 model_training_utils.py:505] Train Step: 789/2100 / loss = 1.298828125 I0421 10:47:14.607496 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.72 examples/second between steps 1488 and 1489 I0421 10:47:15.698932 47076539613632 model_training_utils.py:505] Train Step: 790/2100 / loss = 1.250732421875 I0421 10:47:15.699373 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.10 examples/second between steps 1489 and 1490 I0421 10:47:16.786372 47076539613632 model_training_utils.py:505] Train Step: 791/2100 / loss = 1.22998046875 I0421 10:47:16.786799 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.61 examples/second between steps 1490 and 1491 I0421 10:47:17.871682 47076539613632 model_training_utils.py:505] Train Step: 792/2100 / loss = 0.8486328125 I0421 10:47:17.872104 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 1491 and 1492 I0421 10:47:18.959922 47076539613632 model_training_utils.py:505] Train Step: 793/2100 / loss = 0.9205322265625 I0421 10:47:18.960347 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.54 examples/second between steps 1492 and 1493 I0421 10:47:20.054541 47076539613632 model_training_utils.py:505] Train Step: 794/2100 / loss = 1.009521484375 I0421 10:47:20.054936 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.96 examples/second between steps 1493 and 1494 I0421 10:47:21.151968 47076539613632 model_training_utils.py:505] Train Step: 795/2100 / loss = 1.1455078125 I0421 10:47:21.152375 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.53 examples/second between steps 1494 and 1495 I0421 10:47:22.242842 47076539613632 model_training_utils.py:505] Train Step: 796/2100 / loss = 1.2252197265625 I0421 10:47:22.243219 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.26 examples/second between steps 1495 and 1496 I0421 10:47:23.329753 47076539613632 model_training_utils.py:505] Train Step: 797/2100 / loss = 1.03857421875 I0421 10:47:23.330134 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 1496 and 1497 I0421 10:47:24.419654 47076539613632 model_training_utils.py:505] Train Step: 798/2100 / loss = 1.027099609375 I0421 10:47:24.420046 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.30 examples/second between steps 1497 and 1498 I0421 10:47:25.508262 47076539613632 model_training_utils.py:505] Train Step: 799/2100 / loss = 1.1103515625 I0421 10:47:25.508653 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.46 examples/second between steps 1498 and 1499 I0421 10:47:26.603827 47076539613632 model_training_utils.py:505] Train Step: 800/2100 / loss = 1.146240234375 I0421 10:47:26.604221 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.72 examples/second between steps 1499 and 1500 I0421 10:47:27.702528 47076539613632 model_training_utils.py:505] Train Step: 801/2100 / loss = 1.175048828125 I0421 10:47:27.702913 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.38 examples/second between steps 1500 and 1501 I0421 10:47:28.795742 47076539613632 model_training_utils.py:505] Train Step: 802/2100 / loss = 0.9207763671875 I0421 10:47:28.796141 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.93 examples/second between steps 1501 and 1502 I0421 10:47:29.894442 47076539613632 model_training_utils.py:505] Train Step: 803/2100 / loss = 0.945068359375 I0421 10:47:29.894822 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.39 examples/second between steps 1502 and 1503 I0421 10:47:30.993613 47076539613632 model_training_utils.py:505] Train Step: 804/2100 / loss = 0.962890625 I0421 10:47:30.993997 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.31 examples/second between steps 1503 and 1504 I0421 10:47:32.094469 47076539613632 model_training_utils.py:505] Train Step: 805/2100 / loss = 1.44775390625 I0421 10:47:32.094856 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.15 examples/second between steps 1504 and 1505 I0421 10:47:33.191532 47076539613632 model_training_utils.py:505] Train Step: 806/2100 / loss = 1.20703125 I0421 10:47:33.191913 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.59 examples/second between steps 1505 and 1506 I0421 10:47:34.287504 47076539613632 model_training_utils.py:505] Train Step: 807/2100 / loss = 0.794921875 I0421 10:47:34.287894 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.67 examples/second between steps 1506 and 1507 I0421 10:47:35.376121 47076539613632 model_training_utils.py:505] Train Step: 808/2100 / loss = 0.9906005859375 I0421 10:47:35.376517 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.46 examples/second between steps 1507 and 1508 I0421 10:47:36.463880 47076539613632 model_training_utils.py:505] Train Step: 809/2100 / loss = 1.118408203125 I0421 10:47:36.464266 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.54 examples/second between steps 1508 and 1509 I0421 10:47:37.555967 47076539613632 model_training_utils.py:505] Train Step: 810/2100 / loss = 1.2586669921875 I0421 10:47:37.556366 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.08 examples/second between steps 1509 and 1510 I0421 10:47:38.639316 47076539613632 model_training_utils.py:505] Train Step: 811/2100 / loss = 1.19140625 I0421 10:47:38.639701 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 1510 and 1511 I0421 10:47:39.731784 47076539613632 model_training_utils.py:505] Train Step: 812/2100 / loss = 1.32568359375 I0421 10:47:39.732174 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.05 examples/second between steps 1511 and 1512 I0421 10:47:40.822713 47076539613632 model_training_utils.py:505] Train Step: 813/2100 / loss = 1.3233642578125 I0421 10:47:40.823099 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.22 examples/second between steps 1512 and 1513 I0421 10:47:41.915500 47076539613632 model_training_utils.py:505] Train Step: 814/2100 / loss = 0.9251708984375 I0421 10:47:41.915892 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.02 examples/second between steps 1513 and 1514 I0421 10:47:43.011878 47076539613632 model_training_utils.py:505] Train Step: 815/2100 / loss = 0.8245849609375 I0421 10:47:43.012268 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.64 examples/second between steps 1514 and 1515 I0421 10:47:44.106042 47076539613632 model_training_utils.py:505] Train Step: 816/2100 / loss = 1.1522216796875 I0421 10:47:44.106439 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.86 examples/second between steps 1515 and 1516 I0421 10:47:45.200642 47076539613632 model_training_utils.py:505] Train Step: 817/2100 / loss = 1.512451171875 I0421 10:47:45.201024 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.83 examples/second between steps 1516 and 1517 I0421 10:47:46.298630 47076539613632 model_training_utils.py:505] Train Step: 818/2100 / loss = 1.1314697265625 I0421 10:47:46.299017 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.44 examples/second between steps 1517 and 1518 I0421 10:47:47.383451 47076539613632 model_training_utils.py:505] Train Step: 819/2100 / loss = 0.888671875 I0421 10:47:47.383831 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 1518 and 1519 I0421 10:47:48.471593 47076539613632 model_training_utils.py:505] Train Step: 820/2100 / loss = 0.947509765625 I0421 10:47:48.471983 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.49 examples/second between steps 1519 and 1520 I0421 10:47:49.557766 47076539613632 model_training_utils.py:505] Train Step: 821/2100 / loss = 1.314453125 I0421 10:47:49.558157 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 1520 and 1521 I0421 10:47:50.660841 47076539613632 model_training_utils.py:505] Train Step: 822/2100 / loss = 1.2099609375 I0421 10:47:50.661225 47076539613632 keras_utils.py:133] TimeHistory: 1.10 seconds, 116.88 examples/second between steps 1521 and 1522 I0421 10:47:51.756274 47076539613632 model_training_utils.py:505] Train Step: 823/2100 / loss = 1.46630859375 I0421 10:47:51.756664 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.74 examples/second between steps 1522 and 1523 I0421 10:47:52.850459 47076539613632 model_training_utils.py:505] Train Step: 824/2100 / loss = 1.344482421875 I0421 10:47:52.850841 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.89 examples/second between steps 1523 and 1524 I0421 10:47:53.952278 47076539613632 model_training_utils.py:505] Train Step: 825/2100 / loss = 1.1978759765625 I0421 10:47:53.952715 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.04 examples/second between steps 1524 and 1525 I0421 10:47:55.053365 47076539613632 model_training_utils.py:505] Train Step: 826/2100 / loss = 1.2626953125 I0421 10:47:55.053791 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.14 examples/second between steps 1525 and 1526 I0421 10:47:56.153141 47076539613632 model_training_utils.py:505] Train Step: 827/2100 / loss = 1.35302734375 I0421 10:47:56.153578 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.29 examples/second between steps 1526 and 1527 I0421 10:47:57.248036 47076539613632 model_training_utils.py:505] Train Step: 828/2100 / loss = 1.533203125 I0421 10:47:57.248470 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.79 examples/second between steps 1527 and 1528 I0421 10:47:58.340044 47076539613632 model_training_utils.py:505] Train Step: 829/2100 / loss = 1.324951171875 I0421 10:47:58.340475 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.11 examples/second between steps 1528 and 1529 I0421 10:47:59.431691 47076539613632 model_training_utils.py:505] Train Step: 830/2100 / loss = 1.02099609375 I0421 10:47:59.432129 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.13 examples/second between steps 1529 and 1530 I0421 10:48:00.520510 47076539613632 model_training_utils.py:505] Train Step: 831/2100 / loss = 1.1087646484375 I0421 10:48:00.520941 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.49 examples/second between steps 1530 and 1531 I0421 10:48:01.611633 47076539613632 model_training_utils.py:505] Train Step: 832/2100 / loss = 1.18603515625 I0421 10:48:01.612056 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.20 examples/second between steps 1531 and 1532 I0421 10:48:02.705230 47076539613632 model_training_utils.py:505] Train Step: 833/2100 / loss = 0.8897705078125 I0421 10:48:02.705661 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.91 examples/second between steps 1532 and 1533 I0421 10:48:03.789345 47076539613632 model_training_utils.py:505] Train Step: 834/2100 / loss = 1.03515625 I0421 10:48:03.789772 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 1533 and 1534 I0421 10:48:04.874679 47076539613632 model_training_utils.py:505] Train Step: 835/2100 / loss = 1.40771484375 I0421 10:48:04.875126 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 1534 and 1535 I0421 10:48:05.966503 47076539613632 model_training_utils.py:505] Train Step: 836/2100 / loss = 1.1923828125 I0421 10:48:05.966930 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.10 examples/second between steps 1535 and 1536 I0421 10:48:07.062394 47076539613632 model_training_utils.py:505] Train Step: 837/2100 / loss = 1.24267578125 I0421 10:48:07.062819 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.68 examples/second between steps 1536 and 1537 I0421 10:48:08.155940 47076539613632 model_training_utils.py:505] Train Step: 838/2100 / loss = 1.178466796875 I0421 10:48:08.156367 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.93 examples/second between steps 1537 and 1538 I0421 10:48:09.248010 47076539613632 model_training_utils.py:505] Train Step: 839/2100 / loss = 1.244384765625 I0421 10:48:09.248450 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.10 examples/second between steps 1538 and 1539 I0421 10:48:10.340926 47076539613632 model_training_utils.py:505] Train Step: 840/2100 / loss = 1.081787109375 I0421 10:48:10.341366 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.00 examples/second between steps 1539 and 1540 I0421 10:48:11.431592 47076539613632 model_training_utils.py:505] Train Step: 841/2100 / loss = 1.2496337890625 I0421 10:48:11.432014 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.25 examples/second between steps 1540 and 1541 I0421 10:48:12.527532 47076539613632 model_training_utils.py:505] Train Step: 842/2100 / loss = 1.407470703125 I0421 10:48:12.527962 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.65 examples/second between steps 1541 and 1542 I0421 10:48:13.623888 47076539613632 model_training_utils.py:505] Train Step: 843/2100 / loss = 1.057373046875 I0421 10:48:13.624328 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.59 examples/second between steps 1542 and 1543 I0421 10:48:14.719745 47076539613632 model_training_utils.py:505] Train Step: 844/2100 / loss = 1.071044921875 I0421 10:48:14.720177 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.67 examples/second between steps 1543 and 1544 I0421 10:48:15.811988 47076539613632 model_training_utils.py:505] Train Step: 845/2100 / loss = 1.109619140625 I0421 10:48:15.812424 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.07 examples/second between steps 1544 and 1545 I0421 10:48:16.899119 47076539613632 model_training_utils.py:505] Train Step: 846/2100 / loss = 1.29052734375 I0421 10:48:16.899554 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 1545 and 1546 I0421 10:48:17.983568 47076539613632 model_training_utils.py:505] Train Step: 847/2100 / loss = 0.9256591796875 I0421 10:48:17.984001 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 1546 and 1547 I0421 10:48:19.070128 47076539613632 model_training_utils.py:505] Train Step: 848/2100 / loss = 1.114990234375 I0421 10:48:19.070564 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.70 examples/second between steps 1547 and 1548 I0421 10:48:20.158898 47076539613632 model_training_utils.py:505] Train Step: 849/2100 / loss = 0.916015625 I0421 10:48:20.159340 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.47 examples/second between steps 1548 and 1549 I0421 10:48:21.245922 47076539613632 model_training_utils.py:505] Train Step: 850/2100 / loss = 0.78515625 I0421 10:48:21.246354 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 1549 and 1550 I0421 10:48:22.339018 47076539613632 model_training_utils.py:505] Train Step: 851/2100 / loss = 0.63055419921875 I0421 10:48:22.339456 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.01 examples/second between steps 1550 and 1551 I0421 10:48:23.431870 47076539613632 model_training_utils.py:505] Train Step: 852/2100 / loss = 0.6890869140625 I0421 10:48:23.432317 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.00 examples/second between steps 1551 and 1552 I0421 10:48:24.522451 47076539613632 model_training_utils.py:505] Train Step: 853/2100 / loss = 0.7177734375 I0421 10:48:24.522895 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.27 examples/second between steps 1552 and 1553 I0421 10:48:25.624447 47076539613632 model_training_utils.py:505] Train Step: 854/2100 / loss = 0.8251953125 I0421 10:48:25.624876 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.04 examples/second between steps 1553 and 1554 I0421 10:48:26.725620 47076539613632 model_training_utils.py:505] Train Step: 855/2100 / loss = 1.123291015625 I0421 10:48:26.726048 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.12 examples/second between steps 1554 and 1555 I0421 10:48:27.820781 47076539613632 model_training_utils.py:505] Train Step: 856/2100 / loss = 1.2197265625 I0421 10:48:27.821206 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.76 examples/second between steps 1555 and 1556 I0421 10:48:28.903899 47076539613632 model_training_utils.py:505] Train Step: 857/2100 / loss = 1.133544921875 I0421 10:48:28.904336 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 1556 and 1557 I0421 10:48:29.988299 47076539613632 model_training_utils.py:505] Train Step: 858/2100 / loss = 1.37744140625 I0421 10:48:29.988727 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 1557 and 1558 I0421 10:48:31.081651 47076539613632 model_training_utils.py:505] Train Step: 859/2100 / loss = 0.7958984375 I0421 10:48:31.082075 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.94 examples/second between steps 1558 and 1559 I0421 10:48:32.174510 47076539613632 model_training_utils.py:505] Train Step: 860/2100 / loss = 0.8917236328125 I0421 10:48:32.174938 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.00 examples/second between steps 1559 and 1560 I0421 10:48:33.270327 47076539613632 model_training_utils.py:505] Train Step: 861/2100 / loss = 0.9869384765625 I0421 10:48:33.270752 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.68 examples/second between steps 1560 and 1561 I0421 10:48:34.358778 47076539613632 model_training_utils.py:505] Train Step: 862/2100 / loss = 1.639404296875 I0421 10:48:34.359209 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.47 examples/second between steps 1561 and 1562 I0421 10:48:35.449362 47076539613632 model_training_utils.py:505] Train Step: 863/2100 / loss = 1.591796875 I0421 10:48:35.449791 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.23 examples/second between steps 1562 and 1563 I0421 10:48:36.529512 47076539613632 model_training_utils.py:505] Train Step: 864/2100 / loss = 1.1722412109375 I0421 10:48:36.529947 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.39 examples/second between steps 1563 and 1564 I0421 10:48:37.613673 47076539613632 model_training_utils.py:505] Train Step: 865/2100 / loss = 1.0703125 I0421 10:48:37.614099 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 1564 and 1565 I0421 10:48:38.699271 47076539613632 model_training_utils.py:505] Train Step: 866/2100 / loss = 0.7266845703125 I0421 10:48:38.699699 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 1565 and 1566 I0421 10:48:39.782126 47076539613632 model_training_utils.py:505] Train Step: 867/2100 / loss = 0.77685546875 I0421 10:48:39.782563 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 1566 and 1567 I0421 10:48:40.866231 47076539613632 model_training_utils.py:505] Train Step: 868/2100 / loss = 1.2529296875 I0421 10:48:40.866668 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 1567 and 1568 I0421 10:48:41.951580 47076539613632 model_training_utils.py:505] Train Step: 869/2100 / loss = 0.9278564453125 I0421 10:48:41.952001 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 1568 and 1569 I0421 10:48:43.044303 47076539613632 model_training_utils.py:505] Train Step: 870/2100 / loss = 0.7664794921875 I0421 10:48:43.044726 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.02 examples/second between steps 1569 and 1570 I0421 10:48:44.142018 47076539613632 model_training_utils.py:505] Train Step: 871/2100 / loss = 0.9378662109375 I0421 10:48:44.142464 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.46 examples/second between steps 1570 and 1571 I0421 10:48:45.229272 47076539613632 model_training_utils.py:505] Train Step: 872/2100 / loss = 0.849853515625 I0421 10:48:45.229706 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.59 examples/second between steps 1571 and 1572 I0421 10:48:46.314747 47076539613632 model_training_utils.py:505] Train Step: 873/2100 / loss = 1.0556640625 I0421 10:48:46.315169 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 1572 and 1573 I0421 10:48:47.404043 47076539613632 model_training_utils.py:505] Train Step: 874/2100 / loss = 1.154541015625 I0421 10:48:47.404481 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.42 examples/second between steps 1573 and 1574 I0421 10:48:48.491350 47076539613632 model_training_utils.py:505] Train Step: 875/2100 / loss = 0.89599609375 I0421 10:48:48.491779 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.59 examples/second between steps 1574 and 1575 I0421 10:48:49.582478 47076539613632 model_training_utils.py:505] Train Step: 876/2100 / loss = 0.9085693359375 I0421 10:48:49.582912 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.22 examples/second between steps 1575 and 1576 I0421 10:48:50.666463 47076539613632 model_training_utils.py:505] Train Step: 877/2100 / loss = 1.00927734375 I0421 10:48:50.666887 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 1576 and 1577 I0421 10:48:51.760179 47076539613632 model_training_utils.py:505] Train Step: 878/2100 / loss = 0.743896484375 I0421 10:48:51.760604 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.92 examples/second between steps 1577 and 1578 I0421 10:48:52.855700 47076539613632 model_training_utils.py:505] Train Step: 879/2100 / loss = 0.7587890625 I0421 10:48:52.856122 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.73 examples/second between steps 1578 and 1579 I0421 10:48:53.953251 47076539613632 model_training_utils.py:505] Train Step: 880/2100 / loss = 1.10009765625 I0421 10:48:53.953683 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.47 examples/second between steps 1579 and 1580 I0421 10:48:55.047940 47076539613632 model_training_utils.py:505] Train Step: 881/2100 / loss = 1.44580078125 I0421 10:48:55.048375 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.79 examples/second between steps 1580 and 1581 I0421 10:48:56.148607 47076539613632 model_training_utils.py:505] Train Step: 882/2100 / loss = 1.35546875 I0421 10:48:56.149033 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.15 examples/second between steps 1581 and 1582 I0421 10:48:57.245627 47076539613632 model_training_utils.py:505] Train Step: 883/2100 / loss = 1.0546875 I0421 10:48:57.246049 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.54 examples/second between steps 1582 and 1583 I0421 10:48:58.334143 47076539613632 model_training_utils.py:505] Train Step: 884/2100 / loss = 1.01513671875 I0421 10:48:58.334571 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.47 examples/second between steps 1583 and 1584 I0421 10:48:59.425756 47076539613632 model_training_utils.py:505] Train Step: 885/2100 / loss = 0.747802734375 I0421 10:48:59.426184 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.11 examples/second between steps 1584 and 1585 I0421 10:49:00.509982 47076539613632 model_training_utils.py:505] Train Step: 886/2100 / loss = 0.9036865234375 I0421 10:49:00.510429 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 1585 and 1586 I0421 10:49:01.591265 47076539613632 model_training_utils.py:505] Train Step: 887/2100 / loss = 0.7894287109375 I0421 10:49:01.591700 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.25 examples/second between steps 1586 and 1587 I0421 10:49:02.672419 47076539613632 model_training_utils.py:505] Train Step: 888/2100 / loss = 0.747314453125 I0421 10:49:02.672846 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.25 examples/second between steps 1587 and 1588 I0421 10:49:03.753845 47076539613632 model_training_utils.py:505] Train Step: 889/2100 / loss = 0.908447265625 I0421 10:49:03.754272 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 1588 and 1589 I0421 10:49:04.843925 47076539613632 model_training_utils.py:505] Train Step: 890/2100 / loss = 0.8060302734375 I0421 10:49:04.844352 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.29 examples/second between steps 1589 and 1590 I0421 10:49:05.938513 47076539613632 model_training_utils.py:505] Train Step: 891/2100 / loss = 1.542236328125 I0421 10:49:05.938939 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.79 examples/second between steps 1590 and 1591 I0421 10:49:07.028554 47076539613632 model_training_utils.py:505] Train Step: 892/2100 / loss = 1.3135986328125 I0421 10:49:07.028977 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.29 examples/second between steps 1591 and 1592 I0421 10:49:08.116142 47076539613632 model_training_utils.py:505] Train Step: 893/2100 / loss = 1.79638671875 I0421 10:49:08.116572 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 1592 and 1593 I0421 10:49:09.204869 47076539613632 model_training_utils.py:505] Train Step: 894/2100 / loss = 1.074462890625 I0421 10:49:09.205304 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.44 examples/second between steps 1593 and 1594 I0421 10:49:10.292535 47076539613632 model_training_utils.py:505] Train Step: 895/2100 / loss = 0.8701171875 I0421 10:49:10.292961 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.56 examples/second between steps 1594 and 1595 I0421 10:49:11.380311 47076539613632 model_training_utils.py:505] Train Step: 896/2100 / loss = 0.9775390625 I0421 10:49:11.380740 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.58 examples/second between steps 1595 and 1596 I0421 10:49:12.470787 47076539613632 model_training_utils.py:505] Train Step: 897/2100 / loss = 0.7843017578125 I0421 10:49:12.471214 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.25 examples/second between steps 1596 and 1597 I0421 10:49:13.558257 47076539613632 model_training_utils.py:505] Train Step: 898/2100 / loss = 0.9503173828125 I0421 10:49:13.558689 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.57 examples/second between steps 1597 and 1598 I0421 10:49:14.651753 47076539613632 model_training_utils.py:505] Train Step: 899/2100 / loss = 1.244140625 I0421 10:49:14.652183 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.92 examples/second between steps 1598 and 1599 I0421 10:49:15.745344 47076539613632 model_training_utils.py:505] Train Step: 900/2100 / loss = 1.05419921875 I0421 10:49:15.745760 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.91 examples/second between steps 1599 and 1600 I0421 10:49:16.832740 47076539613632 model_training_utils.py:505] Train Step: 901/2100 / loss = 0.8095703125 I0421 10:49:16.833168 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.59 examples/second between steps 1600 and 1601 I0421 10:49:17.924072 47076539613632 model_training_utils.py:505] Train Step: 902/2100 / loss = 0.985595703125 I0421 10:49:17.924508 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.17 examples/second between steps 1601 and 1602 I0421 10:49:19.018221 47076539613632 model_training_utils.py:505] Train Step: 903/2100 / loss = 1.0374755859375 I0421 10:49:19.018672 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.88 examples/second between steps 1602 and 1603 I0421 10:49:20.110978 47076539613632 model_training_utils.py:505] Train Step: 904/2100 / loss = 1.056396484375 I0421 10:49:20.111413 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.15 examples/second between steps 1603 and 1604 I0421 10:49:21.203457 47076539613632 model_training_utils.py:505] Train Step: 905/2100 / loss = 1.09228515625 I0421 10:49:21.203882 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.04 examples/second between steps 1604 and 1605 I0421 10:49:22.294260 47076539613632 model_training_utils.py:505] Train Step: 906/2100 / loss = 0.9730224609375 I0421 10:49:22.294688 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.21 examples/second between steps 1605 and 1606 I0421 10:49:23.375030 47076539613632 model_training_utils.py:505] Train Step: 907/2100 / loss = 1.12451171875 I0421 10:49:23.375464 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.32 examples/second between steps 1606 and 1607 I0421 10:49:24.456682 47076539613632 model_training_utils.py:505] Train Step: 908/2100 / loss = 0.8765869140625 I0421 10:49:24.457112 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 1607 and 1608 I0421 10:49:25.538201 47076539613632 model_training_utils.py:505] Train Step: 909/2100 / loss = 0.67626953125 I0421 10:49:25.538611 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 1608 and 1609 I0421 10:49:26.621787 47076539613632 model_training_utils.py:505] Train Step: 910/2100 / loss = 0.7398681640625 I0421 10:49:26.622211 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 1609 and 1610 I0421 10:49:27.707064 47076539613632 model_training_utils.py:505] Train Step: 911/2100 / loss = 0.7781982421875 I0421 10:49:27.707495 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 1610 and 1611 I0421 10:49:28.793417 47076539613632 model_training_utils.py:505] Train Step: 912/2100 / loss = 0.99365234375 I0421 10:49:28.793837 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.69 examples/second between steps 1611 and 1612 I0421 10:49:29.881301 47076539613632 model_training_utils.py:505] Train Step: 913/2100 / loss = 1.049072265625 I0421 10:49:29.881727 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.53 examples/second between steps 1612 and 1613 I0421 10:49:30.964700 47076539613632 model_training_utils.py:505] Train Step: 914/2100 / loss = 1.099609375 I0421 10:49:30.965138 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 1613 and 1614 I0421 10:49:32.046010 47076539613632 model_training_utils.py:505] Train Step: 915/2100 / loss = 1.20849609375 I0421 10:49:32.046448 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 1614 and 1615 I0421 10:49:33.127905 47076539613632 model_training_utils.py:505] Train Step: 916/2100 / loss = 1.0286865234375 I0421 10:49:33.128337 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 1615 and 1616 I0421 10:49:34.208977 47076539613632 model_training_utils.py:505] Train Step: 917/2100 / loss = 0.8787841796875 I0421 10:49:34.209401 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 1616 and 1617 I0421 10:49:35.290633 47076539613632 model_training_utils.py:505] Train Step: 918/2100 / loss = 0.989501953125 I0421 10:49:35.291061 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 1617 and 1618 I0421 10:49:36.371509 47076539613632 model_training_utils.py:505] Train Step: 919/2100 / loss = 0.96435546875 I0421 10:49:36.371934 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 1618 and 1619 I0421 10:49:37.453422 47076539613632 model_training_utils.py:505] Train Step: 920/2100 / loss = 1.028076171875 I0421 10:49:37.453850 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 1619 and 1620 I0421 10:49:38.536029 47076539613632 model_training_utils.py:505] Train Step: 921/2100 / loss = 0.926025390625 I0421 10:49:38.536465 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 1620 and 1621 I0421 10:49:39.616574 47076539613632 model_training_utils.py:505] Train Step: 922/2100 / loss = 0.848876953125 I0421 10:49:39.617001 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.34 examples/second between steps 1621 and 1622 I0421 10:49:40.701636 47076539613632 model_training_utils.py:505] Train Step: 923/2100 / loss = 1.1561279296875 I0421 10:49:40.702064 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 1622 and 1623 I0421 10:49:41.782395 47076539613632 model_training_utils.py:505] Train Step: 924/2100 / loss = 1.016845703125 I0421 10:49:41.782826 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.31 examples/second between steps 1623 and 1624 I0421 10:49:42.864112 47076539613632 model_training_utils.py:505] Train Step: 925/2100 / loss = 0.8126220703125 I0421 10:49:42.864537 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 1624 and 1625 I0421 10:49:43.945129 47076539613632 model_training_utils.py:505] Train Step: 926/2100 / loss = 0.6988525390625 I0421 10:49:43.945564 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 1625 and 1626 I0421 10:49:45.026015 47076539613632 model_training_utils.py:505] Train Step: 927/2100 / loss = 0.796142578125 I0421 10:49:45.026453 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.29 examples/second between steps 1626 and 1627 I0421 10:49:46.110161 47076539613632 model_training_utils.py:505] Train Step: 928/2100 / loss = 1.037841796875 I0421 10:49:46.110596 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 1627 and 1628 I0421 10:49:47.195786 47076539613632 model_training_utils.py:505] Train Step: 929/2100 / loss = 0.8160400390625 I0421 10:49:47.196216 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 1628 and 1629 I0421 10:49:48.272788 47076539613632 model_training_utils.py:505] Train Step: 930/2100 / loss = 1.12060546875 I0421 10:49:48.273216 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.72 examples/second between steps 1629 and 1630 I0421 10:49:49.356182 47076539613632 model_training_utils.py:505] Train Step: 931/2100 / loss = 0.96630859375 I0421 10:49:49.356624 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 1630 and 1631 I0421 10:49:50.438325 47076539613632 model_training_utils.py:505] Train Step: 932/2100 / loss = 0.84326171875 I0421 10:49:50.438754 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 1631 and 1632 I0421 10:49:51.522360 47076539613632 model_training_utils.py:505] Train Step: 933/2100 / loss = 1.087890625 I0421 10:49:51.522788 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 1632 and 1633 I0421 10:49:52.605834 47076539613632 model_training_utils.py:505] Train Step: 934/2100 / loss = 1.19482421875 I0421 10:49:52.606256 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 1633 and 1634 I0421 10:49:53.690090 47076539613632 model_training_utils.py:505] Train Step: 935/2100 / loss = 1.304443359375 I0421 10:49:53.690522 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 1634 and 1635 I0421 10:49:54.772022 47076539613632 model_training_utils.py:505] Train Step: 936/2100 / loss = 1.41259765625 I0421 10:49:54.772454 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 1635 and 1636 I0421 10:49:55.853628 47076539613632 model_training_utils.py:505] Train Step: 937/2100 / loss = 1.377685546875 I0421 10:49:55.854056 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 1636 and 1637 I0421 10:49:56.933461 47076539613632 model_training_utils.py:505] Train Step: 938/2100 / loss = 1.0550537109375 I0421 10:49:56.933882 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.41 examples/second between steps 1637 and 1638 I0421 10:49:58.014112 47076539613632 model_training_utils.py:505] Train Step: 939/2100 / loss = 0.806640625 I0421 10:49:58.014552 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.29 examples/second between steps 1638 and 1639 I0421 10:49:59.096117 47076539613632 model_training_utils.py:505] Train Step: 940/2100 / loss = 1.09228515625 I0421 10:49:59.096555 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 1639 and 1640 I0421 10:50:00.178165 47076539613632 model_training_utils.py:505] Train Step: 941/2100 / loss = 1.142578125 I0421 10:50:00.178604 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 1640 and 1641 I0421 10:50:01.258883 47076539613632 model_training_utils.py:505] Train Step: 942/2100 / loss = 1.391845703125 I0421 10:50:01.259319 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 1641 and 1642 I0421 10:50:02.337626 47076539613632 model_training_utils.py:505] Train Step: 943/2100 / loss = 1.436279296875 I0421 10:50:02.338054 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.52 examples/second between steps 1642 and 1643 I0421 10:50:03.421120 47076539613632 model_training_utils.py:505] Train Step: 944/2100 / loss = 1.5224609375 I0421 10:50:03.421554 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 1643 and 1644 I0421 10:50:04.504197 47076539613632 model_training_utils.py:505] Train Step: 945/2100 / loss = 1.319091796875 I0421 10:50:04.504636 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 1644 and 1645 I0421 10:50:05.586376 47076539613632 model_training_utils.py:505] Train Step: 946/2100 / loss = 1.68310546875 I0421 10:50:05.586799 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 1645 and 1646 I0421 10:50:06.669510 47076539613632 model_training_utils.py:505] Train Step: 947/2100 / loss = 1.43310546875 I0421 10:50:06.669940 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 1646 and 1647 I0421 10:50:07.749755 47076539613632 model_training_utils.py:505] Train Step: 948/2100 / loss = 1.1593017578125 I0421 10:50:07.750185 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.38 examples/second between steps 1647 and 1648 I0421 10:50:08.833479 47076539613632 model_training_utils.py:505] Train Step: 949/2100 / loss = 0.96142578125 I0421 10:50:08.833899 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 1648 and 1649 I0421 10:50:09.916466 47076539613632 model_training_utils.py:505] Train Step: 950/2100 / loss = 0.9580078125 I0421 10:50:09.916893 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 1649 and 1650 I0421 10:50:11.000303 47076539613632 model_training_utils.py:505] Train Step: 951/2100 / loss = 0.875 I0421 10:50:11.000724 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 1650 and 1651 I0421 10:50:12.080341 47076539613632 model_training_utils.py:505] Train Step: 952/2100 / loss = 0.7005615234375 I0421 10:50:12.080768 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.37 examples/second between steps 1651 and 1652 I0421 10:50:13.160694 47076539613632 model_training_utils.py:505] Train Step: 953/2100 / loss = 0.6920166015625 I0421 10:50:13.161120 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.34 examples/second between steps 1652 and 1653 I0421 10:50:14.240722 47076539613632 model_training_utils.py:505] Train Step: 954/2100 / loss = 0.51190185546875 I0421 10:50:14.241161 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.45 examples/second between steps 1653 and 1654 I0421 10:50:15.321116 47076539613632 model_training_utils.py:505] Train Step: 955/2100 / loss = 0.7860107421875 I0421 10:50:15.321547 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.36 examples/second between steps 1654 and 1655 I0421 10:50:16.400304 47076539613632 model_training_utils.py:505] Train Step: 956/2100 / loss = 0.839599609375 I0421 10:50:16.400727 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.52 examples/second between steps 1655 and 1656 I0421 10:50:17.482536 47076539613632 model_training_utils.py:505] Train Step: 957/2100 / loss = 0.98388671875 I0421 10:50:17.482961 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 1656 and 1657 I0421 10:50:18.565195 47076539613632 model_training_utils.py:505] Train Step: 958/2100 / loss = 0.762939453125 I0421 10:50:18.565627 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 1657 and 1658 I0421 10:50:19.652214 47076539613632 model_training_utils.py:505] Train Step: 959/2100 / loss = 0.829345703125 I0421 10:50:19.652650 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.62 examples/second between steps 1658 and 1659 I0421 10:50:20.735441 47076539613632 model_training_utils.py:505] Train Step: 960/2100 / loss = 1.0306396484375 I0421 10:50:20.735884 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 1659 and 1660 I0421 10:50:21.816008 47076539613632 model_training_utils.py:505] Train Step: 961/2100 / loss = 0.92193603515625 I0421 10:50:21.816439 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.32 examples/second between steps 1660 and 1661 I0421 10:50:22.897410 47076539613632 model_training_utils.py:505] Train Step: 962/2100 / loss = 0.67236328125 I0421 10:50:22.897835 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 1661 and 1662 I0421 10:50:23.979141 47076539613632 model_training_utils.py:505] Train Step: 963/2100 / loss = 0.52911376953125 I0421 10:50:23.979568 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 1662 and 1663 I0421 10:50:25.064355 47076539613632 model_training_utils.py:505] Train Step: 964/2100 / loss = 0.755859375 I0421 10:50:25.064781 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 1663 and 1664 I0421 10:50:26.148587 47076539613632 model_training_utils.py:505] Train Step: 965/2100 / loss = 0.934326171875 I0421 10:50:26.149006 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 1664 and 1665 I0421 10:50:27.233850 47076539613632 model_training_utils.py:505] Train Step: 966/2100 / loss = 0.9853515625 I0421 10:50:27.234275 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 1665 and 1666 I0421 10:50:28.316853 47076539613632 model_training_utils.py:505] Train Step: 967/2100 / loss = 0.8609619140625 I0421 10:50:28.317276 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 1666 and 1667 I0421 10:50:29.393966 47076539613632 model_training_utils.py:505] Train Step: 968/2100 / loss = 0.926025390625 I0421 10:50:29.394406 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.70 examples/second between steps 1667 and 1668 I0421 10:50:30.478711 47076539613632 model_training_utils.py:505] Train Step: 969/2100 / loss = 0.78582763671875 I0421 10:50:30.479140 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 1668 and 1669 I0421 10:50:31.563418 47076539613632 model_training_utils.py:505] Train Step: 970/2100 / loss = 0.890625 I0421 10:50:31.563840 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 1669 and 1670 I0421 10:50:32.649370 47076539613632 model_training_utils.py:505] Train Step: 971/2100 / loss = 0.9056396484375 I0421 10:50:32.649793 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 1670 and 1671 I0421 10:50:33.731170 47076539613632 model_training_utils.py:505] Train Step: 972/2100 / loss = 1.0938720703125 I0421 10:50:33.731606 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 1671 and 1672 I0421 10:50:34.813492 47076539613632 model_training_utils.py:505] Train Step: 973/2100 / loss = 1.360595703125 I0421 10:50:34.813917 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 1672 and 1673 I0421 10:50:35.892236 47076539613632 model_training_utils.py:505] Train Step: 974/2100 / loss = 0.895263671875 I0421 10:50:35.892669 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.52 examples/second between steps 1673 and 1674 I0421 10:50:36.976005 47076539613632 model_training_utils.py:505] Train Step: 975/2100 / loss = 1.241455078125 I0421 10:50:36.976434 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 1674 and 1675 I0421 10:50:38.060729 47076539613632 model_training_utils.py:505] Train Step: 976/2100 / loss = 2.0986328125 I0421 10:50:38.061155 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 1675 and 1676 I0421 10:50:39.143538 47076539613632 model_training_utils.py:505] Train Step: 977/2100 / loss = 2.18017578125 I0421 10:50:39.143966 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 1676 and 1677 I0421 10:50:40.227341 47076539613632 model_training_utils.py:505] Train Step: 978/2100 / loss = 1.768798828125 I0421 10:50:40.227783 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 1677 and 1678 I0421 10:50:41.310268 47076539613632 model_training_utils.py:505] Train Step: 979/2100 / loss = 1.01171875 I0421 10:50:41.310702 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 1678 and 1679 I0421 10:50:42.392879 47076539613632 model_training_utils.py:505] Train Step: 980/2100 / loss = 0.8768310546875 I0421 10:50:42.393316 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 1679 and 1680 I0421 10:50:43.475100 47076539613632 model_training_utils.py:505] Train Step: 981/2100 / loss = 0.79345703125 I0421 10:50:43.475538 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 1680 and 1681 I0421 10:50:44.557044 47076539613632 model_training_utils.py:505] Train Step: 982/2100 / loss = 0.95166015625 I0421 10:50:44.557487 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 1681 and 1682 I0421 10:50:45.635345 47076539613632 model_training_utils.py:505] Train Step: 983/2100 / loss = 0.942138671875 I0421 10:50:45.635782 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.64 examples/second between steps 1682 and 1683 I0421 10:50:46.718333 47076539613632 model_training_utils.py:505] Train Step: 984/2100 / loss = 1.2236328125 I0421 10:50:46.718757 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 1683 and 1684 I0421 10:50:47.799111 47076539613632 model_training_utils.py:505] Train Step: 985/2100 / loss = 1.515380859375 I0421 10:50:47.799545 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.29 examples/second between steps 1684 and 1685 I0421 10:50:48.881441 47076539613632 model_training_utils.py:505] Train Step: 986/2100 / loss = 2.03564453125 I0421 10:50:48.881865 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 1685 and 1686 I0421 10:50:49.967264 47076539613632 model_training_utils.py:505] Train Step: 987/2100 / loss = 1.43701171875 I0421 10:50:49.967707 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 1686 and 1687 I0421 10:50:51.050479 47076539613632 model_training_utils.py:505] Train Step: 988/2100 / loss = 1.552734375 I0421 10:50:51.050901 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 1687 and 1688 I0421 10:50:52.136704 47076539613632 model_training_utils.py:505] Train Step: 989/2100 / loss = 2.6025390625 I0421 10:50:52.137124 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 1688 and 1689 I0421 10:50:53.221307 47076539613632 model_training_utils.py:505] Train Step: 990/2100 / loss = 2.12255859375 I0421 10:50:53.221732 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 1689 and 1690 I0421 10:50:54.305856 47076539613632 model_training_utils.py:505] Train Step: 991/2100 / loss = 1.636962890625 I0421 10:50:54.306278 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 1690 and 1691 I0421 10:50:55.390104 47076539613632 model_training_utils.py:505] Train Step: 992/2100 / loss = 1.059326171875 I0421 10:50:55.390529 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 1691 and 1692 I0421 10:50:56.475252 47076539613632 model_training_utils.py:505] Train Step: 993/2100 / loss = 1.038818359375 I0421 10:50:56.475676 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 1692 and 1693 I0421 10:50:57.559315 47076539613632 model_training_utils.py:505] Train Step: 994/2100 / loss = 0.897216796875 I0421 10:50:57.559738 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 1693 and 1694 I0421 10:50:58.644575 47076539613632 model_training_utils.py:505] Train Step: 995/2100 / loss = 0.9996337890625 I0421 10:50:58.644999 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 1694 and 1695 I0421 10:50:59.727805 47076539613632 model_training_utils.py:505] Train Step: 996/2100 / loss = 1.0718994140625 I0421 10:50:59.728232 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.07 examples/second between steps 1695 and 1696 I0421 10:51:00.814442 47076539613632 model_training_utils.py:505] Train Step: 997/2100 / loss = 1.2694091796875 I0421 10:51:00.814883 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 1696 and 1697 I0421 10:51:01.899778 47076539613632 model_training_utils.py:505] Train Step: 998/2100 / loss = 1.100341796875 I0421 10:51:01.900202 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 1697 and 1698 I0421 10:51:02.982901 47076539613632 model_training_utils.py:505] Train Step: 999/2100 / loss = 1.0242919921875 I0421 10:51:02.983336 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 1698 and 1699 I0421 10:51:04.067524 47076539613632 model_training_utils.py:505] Train Step: 1000/2100 / loss = 1.10498046875 I0421 10:51:04.067953 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 1699 and 1700 I0421 10:51:05.153020 47076539613632 model_training_utils.py:505] Train Step: 1001/2100 / loss = 1.320556640625 I0421 10:51:05.153452 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 1700 and 1701 I0421 10:51:06.239198 47076539613632 model_training_utils.py:505] Train Step: 1002/2100 / loss = 0.981201171875 I0421 10:51:06.239631 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.69 examples/second between steps 1701 and 1702 I0421 10:51:07.326565 47076539613632 model_training_utils.py:505] Train Step: 1003/2100 / loss = 1.19873046875 I0421 10:51:07.326990 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.58 examples/second between steps 1702 and 1703 I0421 10:51:08.411939 47076539613632 model_training_utils.py:505] Train Step: 1004/2100 / loss = 1.1697998046875 I0421 10:51:08.412369 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 1703 and 1704 I0421 10:51:09.497170 47076539613632 model_training_utils.py:505] Train Step: 1005/2100 / loss = 0.74609375 I0421 10:51:09.497603 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 1704 and 1705 I0421 10:51:10.581172 47076539613632 model_training_utils.py:505] Train Step: 1006/2100 / loss = 0.6502685546875 I0421 10:51:10.581563 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 1705 and 1706 I0421 10:51:11.663341 47076539613632 model_training_utils.py:505] Train Step: 1007/2100 / loss = 0.73590087890625 I0421 10:51:11.663730 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 1706 and 1707 I0421 10:51:12.745577 47076539613632 model_training_utils.py:505] Train Step: 1008/2100 / loss = 0.6044921875 I0421 10:51:12.745971 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 1707 and 1708 I0421 10:51:13.828036 47076539613632 model_training_utils.py:505] Train Step: 1009/2100 / loss = 0.884765625 I0421 10:51:13.828427 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 1708 and 1709 I0421 10:51:14.910793 47076539613632 model_training_utils.py:505] Train Step: 1010/2100 / loss = 0.58123779296875 I0421 10:51:14.911174 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 1709 and 1710 I0421 10:51:15.994237 47076539613632 model_training_utils.py:505] Train Step: 1011/2100 / loss = 0.6090087890625 I0421 10:51:15.994626 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.07 examples/second between steps 1710 and 1711 I0421 10:51:17.076777 47076539613632 model_training_utils.py:505] Train Step: 1012/2100 / loss = 0.5616455078125 I0421 10:51:17.077164 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 1711 and 1712 I0421 10:51:18.160674 47076539613632 model_training_utils.py:505] Train Step: 1013/2100 / loss = 0.7169189453125 I0421 10:51:18.161056 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 1712 and 1713 I0421 10:51:19.244286 47076539613632 model_training_utils.py:505] Train Step: 1014/2100 / loss = 1.1163330078125 I0421 10:51:19.244666 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 1713 and 1714 I0421 10:51:20.328818 47076539613632 model_training_utils.py:505] Train Step: 1015/2100 / loss = 1.4755859375 I0421 10:51:20.329199 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 1714 and 1715 I0421 10:51:21.409914 47076539613632 model_training_utils.py:505] Train Step: 1016/2100 / loss = 1.1051025390625 I0421 10:51:21.410303 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.34 examples/second between steps 1715 and 1716 I0421 10:51:22.492776 47076539613632 model_training_utils.py:505] Train Step: 1017/2100 / loss = 1.215087890625 I0421 10:51:22.493147 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.07 examples/second between steps 1716 and 1717 I0421 10:51:23.576621 47076539613632 model_training_utils.py:505] Train Step: 1018/2100 / loss = 0.656005859375 I0421 10:51:23.577003 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 1717 and 1718 I0421 10:51:24.659298 47076539613632 model_training_utils.py:505] Train Step: 1019/2100 / loss = 0.6356201171875 I0421 10:51:24.659673 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 1718 and 1719 I0421 10:51:25.743432 47076539613632 model_training_utils.py:505] Train Step: 1020/2100 / loss = 2.17138671875 I0421 10:51:25.743807 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 1719 and 1720 I0421 10:51:26.829597 47076539613632 model_training_utils.py:505] Train Step: 1021/2100 / loss = 1.85595703125 I0421 10:51:26.829973 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 1720 and 1721 I0421 10:51:27.919341 47076539613632 model_training_utils.py:505] Train Step: 1022/2100 / loss = 1.146484375 I0421 10:51:27.919723 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.36 examples/second between steps 1721 and 1722 I0421 10:51:29.005639 47076539613632 model_training_utils.py:505] Train Step: 1023/2100 / loss = 1.2119140625 I0421 10:51:29.006020 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 1722 and 1723 I0421 10:51:30.089388 47076539613632 model_training_utils.py:505] Train Step: 1024/2100 / loss = 1.173828125 I0421 10:51:30.089765 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 1723 and 1724 I0421 10:51:31.174606 47076539613632 model_training_utils.py:505] Train Step: 1025/2100 / loss = 1.089599609375 I0421 10:51:31.174982 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 1724 and 1725 I0421 10:51:32.257571 47076539613632 model_training_utils.py:505] Train Step: 1026/2100 / loss = 0.914794921875 I0421 10:51:32.257951 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 1725 and 1726 I0421 10:51:33.343132 47076539613632 model_training_utils.py:505] Train Step: 1027/2100 / loss = 1.466552734375 I0421 10:51:33.343521 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 1726 and 1727 I0421 10:51:34.429981 47076539613632 model_training_utils.py:505] Train Step: 1028/2100 / loss = 0.9962158203125 I0421 10:51:34.430371 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.67 examples/second between steps 1727 and 1728 I0421 10:51:35.515481 47076539613632 model_training_utils.py:505] Train Step: 1029/2100 / loss = 0.85040283203125 I0421 10:51:35.515864 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 1728 and 1729 I0421 10:51:36.598911 47076539613632 model_training_utils.py:505] Train Step: 1030/2100 / loss = 0.61962890625 I0421 10:51:36.599300 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 1729 and 1730 I0421 10:51:37.683639 47076539613632 model_training_utils.py:505] Train Step: 1031/2100 / loss = 1.0960693359375 I0421 10:51:37.684029 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 1730 and 1731 I0421 10:51:38.766670 47076539613632 model_training_utils.py:505] Train Step: 1032/2100 / loss = 1.576171875 I0421 10:51:38.767048 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.07 examples/second between steps 1731 and 1732 I0421 10:51:39.851000 47076539613632 model_training_utils.py:505] Train Step: 1033/2100 / loss = 1.59814453125 I0421 10:51:39.851393 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 1732 and 1733 I0421 10:51:40.934458 47076539613632 model_training_utils.py:505] Train Step: 1034/2100 / loss = 1.2669677734375 I0421 10:51:40.934837 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 1733 and 1734 I0421 10:51:42.026116 47076539613632 model_training_utils.py:505] Train Step: 1035/2100 / loss = 1.179443359375 I0421 10:51:42.026500 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.14 examples/second between steps 1734 and 1735 I0421 10:51:43.110892 47076539613632 model_training_utils.py:505] Train Step: 1036/2100 / loss = 1.338134765625 I0421 10:51:43.111273 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 1735 and 1736 I0421 10:51:44.197267 47076539613632 model_training_utils.py:505] Train Step: 1037/2100 / loss = 0.9947509765625 I0421 10:51:44.197659 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.69 examples/second between steps 1736 and 1737 I0421 10:51:45.282352 47076539613632 model_training_utils.py:505] Train Step: 1038/2100 / loss = 0.8668212890625 I0421 10:51:45.282746 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 1737 and 1738 I0421 10:51:46.365214 47076539613632 model_training_utils.py:505] Train Step: 1039/2100 / loss = 0.840576171875 I0421 10:51:46.365593 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 1738 and 1739 I0421 10:51:47.450001 47076539613632 model_training_utils.py:505] Train Step: 1040/2100 / loss = 0.9522705078125 I0421 10:51:47.450364 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 1739 and 1740 I0421 10:51:48.534406 47076539613632 model_training_utils.py:505] Train Step: 1041/2100 / loss = 1.101318359375 I0421 10:51:48.534781 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 1740 and 1741 I0421 10:51:49.617430 47076539613632 model_training_utils.py:505] Train Step: 1042/2100 / loss = 1.219970703125 I0421 10:51:49.617818 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 1741 and 1742 I0421 10:51:50.699578 47076539613632 model_training_utils.py:505] Train Step: 1043/2100 / loss = 2.00830078125 I0421 10:51:50.699961 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 1742 and 1743 I0421 10:51:51.784542 47076539613632 model_training_utils.py:505] Train Step: 1044/2100 / loss = 1.424072265625 I0421 10:51:51.784923 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 1743 and 1744 I0421 10:51:52.868947 47076539613632 model_training_utils.py:505] Train Step: 1045/2100 / loss = 1.47705078125 I0421 10:51:52.869338 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 1744 and 1745 I0421 10:51:53.954406 47076539613632 model_training_utils.py:505] Train Step: 1046/2100 / loss = 1.361083984375 I0421 10:51:53.954797 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 1745 and 1746 I0421 10:51:55.037413 47076539613632 model_training_utils.py:505] Train Step: 1047/2100 / loss = 1.087890625 I0421 10:51:55.037798 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 1746 and 1747 I0421 10:51:56.122765 47076539613632 model_training_utils.py:505] Train Step: 1048/2100 / loss = 1.230224609375 I0421 10:51:56.123150 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 1747 and 1748 I0421 10:51:57.204694 47076539613632 model_training_utils.py:505] Train Step: 1049/2100 / loss = 1.150390625 I0421 10:51:57.205075 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 1748 and 1749 I0421 10:51:58.284749 47076539613632 model_training_utils.py:505] Train Step: 1050/2100 / loss = 1.067138671875 I0421 10:51:58.285130 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.42 examples/second between steps 1749 and 1750 I0421 10:51:59.367043 47076539613632 model_training_utils.py:505] Train Step: 1051/2100 / loss = 1.177978515625 I0421 10:51:59.367448 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 1750 and 1751 I0421 10:52:00.450703 47076539613632 model_training_utils.py:505] Train Step: 1052/2100 / loss = 1.225830078125 I0421 10:52:00.451081 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 1751 and 1752 I0421 10:52:01.538308 47076539613632 model_training_utils.py:505] Train Step: 1053/2100 / loss = 1.153564453125 I0421 10:52:01.538697 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.59 examples/second between steps 1752 and 1753 I0421 10:52:02.625982 47076539613632 model_training_utils.py:505] Train Step: 1054/2100 / loss = 1.03662109375 I0421 10:52:02.626381 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.57 examples/second between steps 1753 and 1754 I0421 10:52:03.712099 47076539613632 model_training_utils.py:505] Train Step: 1055/2100 / loss = 1.078857421875 I0421 10:52:03.712494 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 1754 and 1755 I0421 10:52:04.797888 47076539613632 model_training_utils.py:505] Train Step: 1056/2100 / loss = 0.9310302734375 I0421 10:52:04.798286 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 1755 and 1756 I0421 10:52:05.881896 47076539613632 model_training_utils.py:505] Train Step: 1057/2100 / loss = 1.137939453125 I0421 10:52:05.882295 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.07 examples/second between steps 1756 and 1757 I0421 10:52:06.968670 47076539613632 model_training_utils.py:505] Train Step: 1058/2100 / loss = 0.833740234375 I0421 10:52:06.969054 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 1757 and 1758 I0421 10:52:08.051957 47076539613632 model_training_utils.py:505] Train Step: 1059/2100 / loss = 1.2125244140625 I0421 10:52:08.052357 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.07 examples/second between steps 1758 and 1759 I0421 10:52:09.139094 47076539613632 model_training_utils.py:505] Train Step: 1060/2100 / loss = 2.357421875 I0421 10:52:09.139485 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 1759 and 1760 I0421 10:52:10.224225 47076539613632 model_training_utils.py:505] Train Step: 1061/2100 / loss = 1.64013671875 I0421 10:52:10.224622 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 1760 and 1761 I0421 10:52:11.307011 47076539613632 model_training_utils.py:505] Train Step: 1062/2100 / loss = 0.9632568359375 I0421 10:52:11.307419 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 1761 and 1762 I0421 10:52:12.388060 47076539613632 model_training_utils.py:505] Train Step: 1063/2100 / loss = 1.0938720703125 I0421 10:52:12.388459 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.33 examples/second between steps 1762 and 1763 I0421 10:52:13.470401 47076539613632 model_training_utils.py:505] Train Step: 1064/2100 / loss = 1.3828125 I0421 10:52:13.470783 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 1763 and 1764 I0421 10:52:14.553249 47076539613632 model_training_utils.py:505] Train Step: 1065/2100 / loss = 1.723876953125 I0421 10:52:14.553651 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 1764 and 1765 I0421 10:52:15.635414 47076539613632 model_training_utils.py:505] Train Step: 1066/2100 / loss = 1.62109375 I0421 10:52:15.635801 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 1765 and 1766 I0421 10:52:16.712410 47076539613632 model_training_utils.py:505] Train Step: 1067/2100 / loss = 1.2958984375 I0421 10:52:16.712814 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.78 examples/second between steps 1766 and 1767 I0421 10:52:17.795919 47076539613632 model_training_utils.py:505] Train Step: 1068/2100 / loss = 1.0223388671875 I0421 10:52:17.796318 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 1767 and 1768 I0421 10:52:18.878324 47076539613632 model_training_utils.py:505] Train Step: 1069/2100 / loss = 0.9698486328125 I0421 10:52:18.878710 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 1768 and 1769 I0421 10:52:19.960448 47076539613632 model_training_utils.py:505] Train Step: 1070/2100 / loss = 1.83642578125 I0421 10:52:19.960835 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 1769 and 1770 I0421 10:52:21.042190 47076539613632 model_training_utils.py:505] Train Step: 1071/2100 / loss = 1.8876953125 I0421 10:52:21.042578 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 1770 and 1771 I0421 10:52:22.123611 47076539613632 model_training_utils.py:505] Train Step: 1072/2100 / loss = 1.060302734375 I0421 10:52:22.123996 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 1771 and 1772 I0421 10:52:23.207539 47076539613632 model_training_utils.py:505] Train Step: 1073/2100 / loss = 0.8692626953125 I0421 10:52:23.207925 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 1772 and 1773 I0421 10:52:24.290155 47076539613632 model_training_utils.py:505] Train Step: 1074/2100 / loss = 1.05908203125 I0421 10:52:24.290554 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 1773 and 1774 I0421 10:52:25.377189 47076539613632 model_training_utils.py:505] Train Step: 1075/2100 / loss = 0.93212890625 I0421 10:52:25.377597 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 1774 and 1775 I0421 10:52:26.463442 47076539613632 model_training_utils.py:505] Train Step: 1076/2100 / loss = 0.99560546875 I0421 10:52:26.463828 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 1775 and 1776 I0421 10:52:27.550392 47076539613632 model_training_utils.py:505] Train Step: 1077/2100 / loss = 0.9700927734375 I0421 10:52:27.550778 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 1776 and 1777 I0421 10:52:28.638356 47076539613632 model_training_utils.py:505] Train Step: 1078/2100 / loss = 1.34521484375 I0421 10:52:28.638740 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.53 examples/second between steps 1777 and 1778 I0421 10:52:29.722595 47076539613632 model_training_utils.py:505] Train Step: 1079/2100 / loss = 2.36376953125 I0421 10:52:29.722984 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 1778 and 1779 I0421 10:52:30.808335 47076539613632 model_training_utils.py:505] Train Step: 1080/2100 / loss = 3.00537109375 I0421 10:52:30.808734 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 1779 and 1780 I0421 10:52:31.891414 47076539613632 model_training_utils.py:505] Train Step: 1081/2100 / loss = 2.58984375 I0421 10:52:31.891805 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 1780 and 1781 I0421 10:52:32.972313 47076539613632 model_training_utils.py:505] Train Step: 1082/2100 / loss = 1.426513671875 I0421 10:52:32.972710 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.30 examples/second between steps 1781 and 1782 I0421 10:52:34.054712 47076539613632 model_training_utils.py:505] Train Step: 1083/2100 / loss = 1.4970703125 I0421 10:52:34.055099 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 1782 and 1783 I0421 10:52:35.140212 47076539613632 model_training_utils.py:505] Train Step: 1084/2100 / loss = 1.171142578125 I0421 10:52:35.140609 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 1783 and 1784 I0421 10:52:36.222819 47076539613632 model_training_utils.py:505] Train Step: 1085/2100 / loss = 0.98095703125 I0421 10:52:36.223203 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 1784 and 1785 I0421 10:52:37.306862 47076539613632 model_training_utils.py:505] Train Step: 1086/2100 / loss = 1.1300048828125 I0421 10:52:37.307248 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 1785 and 1786 I0421 10:52:38.389860 47076539613632 model_training_utils.py:505] Train Step: 1087/2100 / loss = 0.8333740234375 I0421 10:52:38.390242 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 1786 and 1787 I0421 10:52:39.474344 47076539613632 model_training_utils.py:505] Train Step: 1088/2100 / loss = 1.047607421875 I0421 10:52:39.474736 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 1787 and 1788 I0421 10:52:40.556836 47076539613632 model_training_utils.py:505] Train Step: 1089/2100 / loss = 1.78125 I0421 10:52:40.557225 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 1788 and 1789 I0421 10:52:41.637357 47076539613632 model_training_utils.py:505] Train Step: 1090/2100 / loss = 2.31689453125 I0421 10:52:41.637744 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.36 examples/second between steps 1789 and 1790 I0421 10:52:42.720416 47076539613632 model_training_utils.py:505] Train Step: 1091/2100 / loss = 1.498046875 I0421 10:52:42.720808 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 1790 and 1791 I0421 10:52:43.803140 47076539613632 model_training_utils.py:505] Train Step: 1092/2100 / loss = 1.316650390625 I0421 10:52:43.803533 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 1791 and 1792 I0421 10:52:44.887296 47076539613632 model_training_utils.py:505] Train Step: 1093/2100 / loss = 0.761474609375 I0421 10:52:44.887687 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 1792 and 1793 I0421 10:52:45.975326 47076539613632 model_training_utils.py:505] Train Step: 1094/2100 / loss = 0.72216796875 I0421 10:52:45.975712 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.52 examples/second between steps 1793 and 1794 I0421 10:52:47.060620 47076539613632 model_training_utils.py:505] Train Step: 1095/2100 / loss = 0.9659423828125 I0421 10:52:47.061031 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 1794 and 1795 I0421 10:52:48.144860 47076539613632 model_training_utils.py:505] Train Step: 1096/2100 / loss = 1.182861328125 I0421 10:52:48.145255 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 1795 and 1796 I0421 10:52:49.227374 47076539613632 model_training_utils.py:505] Train Step: 1097/2100 / loss = 1.1728515625 I0421 10:52:49.227765 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 1796 and 1797 I0421 10:52:50.311612 47076539613632 model_training_utils.py:505] Train Step: 1098/2100 / loss = 1.02685546875 I0421 10:52:50.311996 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 1797 and 1798 I0421 10:52:51.397253 47076539613632 model_training_utils.py:505] Train Step: 1099/2100 / loss = 0.752197265625 I0421 10:52:51.397650 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 1798 and 1799 I0421 10:52:52.482020 47076539613632 model_training_utils.py:505] Train Step: 1100/2100 / loss = 0.995361328125 I0421 10:52:52.482418 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 1799 and 1800 I0421 10:52:53.567533 47076539613632 model_training_utils.py:505] Train Step: 1101/2100 / loss = 0.927734375 I0421 10:52:53.567924 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 1800 and 1801 I0421 10:52:54.651867 47076539613632 model_training_utils.py:505] Train Step: 1102/2100 / loss = 0.78173828125 I0421 10:52:54.652256 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 1801 and 1802 I0421 10:52:55.734133 47076539613632 model_training_utils.py:505] Train Step: 1103/2100 / loss = 1.1383056640625 I0421 10:52:55.734526 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 1802 and 1803 I0421 10:52:56.818473 47076539613632 model_training_utils.py:505] Train Step: 1104/2100 / loss = 1.0970458984375 I0421 10:52:56.818860 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 1803 and 1804 I0421 10:52:57.902221 47076539613632 model_training_utils.py:505] Train Step: 1105/2100 / loss = 1.056640625 I0421 10:52:57.902612 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 1804 and 1805 I0421 10:52:58.987318 47076539613632 model_training_utils.py:505] Train Step: 1106/2100 / loss = 1.699462890625 I0421 10:52:58.987704 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 1805 and 1806 I0421 10:53:00.069885 47076539613632 model_training_utils.py:505] Train Step: 1107/2100 / loss = 1.397216796875 I0421 10:53:00.070270 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 1806 and 1807 I0421 10:53:01.152350 47076539613632 model_training_utils.py:505] Train Step: 1108/2100 / loss = 1.001708984375 I0421 10:53:01.152738 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 1807 and 1808 I0421 10:53:02.231698 47076539613632 model_training_utils.py:505] Train Step: 1109/2100 / loss = 0.8834228515625 I0421 10:53:02.232089 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.48 examples/second between steps 1808 and 1809 I0421 10:53:03.317346 47076539613632 model_training_utils.py:505] Train Step: 1110/2100 / loss = 1.0003662109375 I0421 10:53:03.317733 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 1809 and 1810 I0421 10:53:04.408085 47076539613632 model_training_utils.py:505] Train Step: 1111/2100 / loss = 1.4697265625 I0421 10:53:04.408478 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.24 examples/second between steps 1810 and 1811 I0421 10:53:05.493815 47076539613632 model_training_utils.py:505] Train Step: 1112/2100 / loss = 1.648681640625 I0421 10:53:05.494205 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 1811 and 1812 I0421 10:53:06.579000 47076539613632 model_training_utils.py:505] Train Step: 1113/2100 / loss = 1.267822265625 I0421 10:53:06.579394 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 1812 and 1813 I0421 10:53:07.661863 47076539613632 model_training_utils.py:505] Train Step: 1114/2100 / loss = 1.16064453125 I0421 10:53:07.662252 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 1813 and 1814 I0421 10:53:08.744575 47076539613632 model_training_utils.py:505] Train Step: 1115/2100 / loss = 1.012451171875 I0421 10:53:08.744960 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 1814 and 1815 I0421 10:53:09.826734 47076539613632 model_training_utils.py:505] Train Step: 1116/2100 / loss = 0.8468017578125 I0421 10:53:09.827118 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 1815 and 1816 I0421 10:53:10.910350 47076539613632 model_training_utils.py:505] Train Step: 1117/2100 / loss = 0.8763427734375 I0421 10:53:10.910756 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 1816 and 1817 I0421 10:53:11.993046 47076539613632 model_training_utils.py:505] Train Step: 1118/2100 / loss = 1.236328125 I0421 10:53:11.993472 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 1817 and 1818 I0421 10:53:13.077557 47076539613632 model_training_utils.py:505] Train Step: 1119/2100 / loss = 1.197265625 I0421 10:53:13.077980 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 1818 and 1819 I0421 10:53:14.162181 47076539613632 model_training_utils.py:505] Train Step: 1120/2100 / loss = 1.1865234375 I0421 10:53:14.162616 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 1819 and 1820 I0421 10:53:15.245908 47076539613632 model_training_utils.py:505] Train Step: 1121/2100 / loss = 1.3134765625 I0421 10:53:15.246340 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 1820 and 1821 I0421 10:53:16.333937 47076539613632 model_training_utils.py:505] Train Step: 1122/2100 / loss = 1.427978515625 I0421 10:53:16.334374 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.52 examples/second between steps 1821 and 1822 I0421 10:53:17.418447 47076539613632 model_training_utils.py:505] Train Step: 1123/2100 / loss = 1.45556640625 I0421 10:53:17.418881 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 1822 and 1823 I0421 10:53:18.502278 47076539613632 model_training_utils.py:505] Train Step: 1124/2100 / loss = 1.177734375 I0421 10:53:18.502713 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 1823 and 1824 I0421 10:53:19.587835 47076539613632 model_training_utils.py:505] Train Step: 1125/2100 / loss = 1.92724609375 I0421 10:53:19.588272 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 1824 and 1825 I0421 10:53:20.675776 47076539613632 model_training_utils.py:505] Train Step: 1126/2100 / loss = 1.6875 I0421 10:53:20.676196 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.57 examples/second between steps 1825 and 1826 I0421 10:53:21.762902 47076539613632 model_training_utils.py:505] Train Step: 1127/2100 / loss = 1.8349609375 I0421 10:53:21.763336 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.63 examples/second between steps 1826 and 1827 I0421 10:53:22.845362 47076539613632 model_training_utils.py:505] Train Step: 1128/2100 / loss = 1.79736328125 I0421 10:53:22.845803 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 1827 and 1828 I0421 10:53:23.930890 47076539613632 model_training_utils.py:505] Train Step: 1129/2100 / loss = 1.86376953125 I0421 10:53:23.931324 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 1828 and 1829 I0421 10:53:25.016458 47076539613632 model_training_utils.py:505] Train Step: 1130/2100 / loss = 1.299072265625 I0421 10:53:25.016880 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 1829 and 1830 I0421 10:53:26.099329 47076539613632 model_training_utils.py:505] Train Step: 1131/2100 / loss = 0.856689453125 I0421 10:53:26.099751 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 1830 and 1831 I0421 10:53:27.180278 47076539613632 model_training_utils.py:505] Train Step: 1132/2100 / loss = 1.084228515625 I0421 10:53:27.180708 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.30 examples/second between steps 1831 and 1832 I0421 10:53:28.263242 47076539613632 model_training_utils.py:505] Train Step: 1133/2100 / loss = 0.9697265625 I0421 10:53:28.263673 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.07 examples/second between steps 1832 and 1833 I0421 10:53:29.345495 47076539613632 model_training_utils.py:505] Train Step: 1134/2100 / loss = 1.232177734375 I0421 10:53:29.345920 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 1833 and 1834 I0421 10:53:30.431170 47076539613632 model_training_utils.py:505] Train Step: 1135/2100 / loss = 1.14453125 I0421 10:53:30.431603 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 1834 and 1835 I0421 10:53:31.515866 47076539613632 model_training_utils.py:505] Train Step: 1136/2100 / loss = 1.233642578125 I0421 10:53:31.516278 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 1835 and 1836 I0421 10:53:32.600317 47076539613632 model_training_utils.py:505] Train Step: 1137/2100 / loss = 1.3662109375 I0421 10:53:32.600718 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 1836 and 1837 I0421 10:53:33.684798 47076539613632 model_training_utils.py:505] Train Step: 1138/2100 / loss = 1.175048828125 I0421 10:53:33.685220 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 1837 and 1838 I0421 10:53:34.768415 47076539613632 model_training_utils.py:505] Train Step: 1139/2100 / loss = 0.8875732421875 I0421 10:53:34.768842 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 1838 and 1839 I0421 10:53:35.851450 47076539613632 model_training_utils.py:505] Train Step: 1140/2100 / loss = 1.2022705078125 I0421 10:53:35.851876 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 1839 and 1840 I0421 10:53:36.933672 47076539613632 model_training_utils.py:505] Train Step: 1141/2100 / loss = 1.736083984375 I0421 10:53:36.934098 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 1840 and 1841 I0421 10:53:38.015786 47076539613632 model_training_utils.py:505] Train Step: 1142/2100 / loss = 2.281005859375 I0421 10:53:38.016221 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 1841 and 1842 I0421 10:53:39.100205 47076539613632 model_training_utils.py:505] Train Step: 1143/2100 / loss = 1.174072265625 I0421 10:53:39.100644 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 1842 and 1843 I0421 10:53:40.185824 47076539613632 model_training_utils.py:505] Train Step: 1144/2100 / loss = 1.438720703125 I0421 10:53:40.186246 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 1843 and 1844 I0421 10:53:41.271600 47076539613632 model_training_utils.py:505] Train Step: 1145/2100 / loss = 1.4921875 I0421 10:53:41.272018 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 1844 and 1845 I0421 10:53:42.356817 47076539613632 model_training_utils.py:505] Train Step: 1146/2100 / loss = 1.074462890625 I0421 10:53:42.357259 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 1845 and 1846 I0421 10:53:43.440479 47076539613632 model_training_utils.py:505] Train Step: 1147/2100 / loss = 0.8280029296875 I0421 10:53:43.440901 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 1846 and 1847 I0421 10:53:44.524798 47076539613632 model_training_utils.py:505] Train Step: 1148/2100 / loss = 0.70703125 I0421 10:53:44.525198 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 1847 and 1848 I0421 10:53:45.607828 47076539613632 model_training_utils.py:505] Train Step: 1149/2100 / loss = 1.03125 I0421 10:53:45.608248 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.07 examples/second between steps 1848 and 1849 I0421 10:53:46.692790 47076539613632 model_training_utils.py:505] Train Step: 1150/2100 / loss = 1.68359375 I0421 10:53:46.693208 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 1849 and 1850 I0421 10:53:47.778216 47076539613632 model_training_utils.py:505] Train Step: 1151/2100 / loss = 1.478271484375 I0421 10:53:47.778651 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 1850 and 1851 I0421 10:53:48.862123 47076539613632 model_training_utils.py:505] Train Step: 1152/2100 / loss = 1.118408203125 I0421 10:53:48.862547 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 1851 and 1852 I0421 10:53:49.946372 47076539613632 model_training_utils.py:505] Train Step: 1153/2100 / loss = 1.367919921875 I0421 10:53:49.946796 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 1852 and 1853 I0421 10:53:51.031107 47076539613632 model_training_utils.py:505] Train Step: 1154/2100 / loss = 1.727294921875 I0421 10:53:51.031543 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 1853 and 1854 I0421 10:53:52.112011 47076539613632 model_training_utils.py:505] Train Step: 1155/2100 / loss = 1.2548828125 I0421 10:53:52.112442 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.30 examples/second between steps 1854 and 1855 I0421 10:53:53.193321 47076539613632 model_training_utils.py:505] Train Step: 1156/2100 / loss = 1.1986083984375 I0421 10:53:53.193749 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.30 examples/second between steps 1855 and 1856 I0421 10:53:54.276914 47076539613632 model_training_utils.py:505] Train Step: 1157/2100 / loss = 1.98828125 I0421 10:53:54.277354 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 1856 and 1857 I0421 10:53:55.352916 47076539613632 model_training_utils.py:505] Train Step: 1158/2100 / loss = 1.986572265625 I0421 10:53:55.353346 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.86 examples/second between steps 1857 and 1858 I0421 10:53:56.435930 47076539613632 model_training_utils.py:505] Train Step: 1159/2100 / loss = 1.88037109375 I0421 10:53:56.436362 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 1858 and 1859 I0421 10:53:57.521652 47076539613632 model_training_utils.py:505] Train Step: 1160/2100 / loss = 1.765869140625 I0421 10:53:57.522065 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 1859 and 1860 I0421 10:53:58.611001 47076539613632 model_training_utils.py:505] Train Step: 1161/2100 / loss = 1.96142578125 I0421 10:53:58.611415 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.41 examples/second between steps 1860 and 1861 I0421 10:53:59.699618 47076539613632 model_training_utils.py:505] Train Step: 1162/2100 / loss = 1.38525390625 I0421 10:53:59.700045 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.46 examples/second between steps 1861 and 1862 I0421 10:54:00.786390 47076539613632 model_training_utils.py:505] Train Step: 1163/2100 / loss = 1.0638427734375 I0421 10:54:00.786814 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.67 examples/second between steps 1862 and 1863 I0421 10:54:01.870579 47076539613632 model_training_utils.py:505] Train Step: 1164/2100 / loss = 1.076904296875 I0421 10:54:01.871008 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 1863 and 1864 I0421 10:54:02.953561 47076539613632 model_training_utils.py:505] Train Step: 1165/2100 / loss = 1.082275390625 I0421 10:54:02.953987 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.07 examples/second between steps 1864 and 1865 I0421 10:54:04.038755 47076539613632 model_training_utils.py:505] Train Step: 1166/2100 / loss = 1.288330078125 I0421 10:54:04.039189 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 1865 and 1866 I0421 10:54:05.121792 47076539613632 model_training_utils.py:505] Train Step: 1167/2100 / loss = 1.136962890625 I0421 10:54:05.122221 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 1866 and 1867 I0421 10:54:06.203123 47076539613632 model_training_utils.py:505] Train Step: 1168/2100 / loss = 1.23779296875 I0421 10:54:06.203557 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 1867 and 1868 I0421 10:54:07.289494 47076539613632 model_training_utils.py:505] Train Step: 1169/2100 / loss = 1.369384765625 I0421 10:54:07.289919 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.67 examples/second between steps 1868 and 1869 I0421 10:54:08.369586 47076539613632 model_training_utils.py:505] Train Step: 1170/2100 / loss = 1.55224609375 I0421 10:54:08.369996 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.38 examples/second between steps 1869 and 1870 I0421 10:54:09.451718 47076539613632 model_training_utils.py:505] Train Step: 1171/2100 / loss = 1.063720703125 I0421 10:54:09.452132 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 1870 and 1871 I0421 10:54:10.535302 47076539613632 model_training_utils.py:505] Train Step: 1172/2100 / loss = 1.0889892578125 I0421 10:54:10.535720 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 1871 and 1872 I0421 10:54:11.618632 47076539613632 model_training_utils.py:505] Train Step: 1173/2100 / loss = 1.13037109375 I0421 10:54:11.619057 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 1872 and 1873 I0421 10:54:12.697803 47076539613632 model_training_utils.py:505] Train Step: 1174/2100 / loss = 1.14404296875 I0421 10:54:12.698228 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.45 examples/second between steps 1873 and 1874 I0421 10:54:13.775733 47076539613632 model_training_utils.py:505] Train Step: 1175/2100 / loss = 1.1214599609375 I0421 10:54:13.776154 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.59 examples/second between steps 1874 and 1875 I0421 10:54:14.854395 47076539613632 model_training_utils.py:505] Train Step: 1176/2100 / loss = 1.1678466796875 I0421 10:54:14.854811 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.51 examples/second between steps 1875 and 1876 I0421 10:54:15.938728 47076539613632 model_training_utils.py:505] Train Step: 1177/2100 / loss = 0.90869140625 I0421 10:54:15.939152 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 1876 and 1877 I0421 10:54:17.021036 47076539613632 model_training_utils.py:505] Train Step: 1178/2100 / loss = 0.9442138671875 I0421 10:54:17.021471 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 1877 and 1878 I0421 10:54:18.105101 47076539613632 model_training_utils.py:505] Train Step: 1179/2100 / loss = 1.094482421875 I0421 10:54:18.105535 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 1878 and 1879 I0421 10:54:19.188184 47076539613632 model_training_utils.py:505] Train Step: 1180/2100 / loss = 1.1021728515625 I0421 10:54:19.188609 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.07 examples/second between steps 1879 and 1880 I0421 10:54:20.272323 47076539613632 model_training_utils.py:505] Train Step: 1181/2100 / loss = 1.15478515625 I0421 10:54:20.272751 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 1880 and 1881 I0421 10:54:21.356954 47076539613632 model_training_utils.py:505] Train Step: 1182/2100 / loss = 0.792724609375 I0421 10:54:21.357385 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 1881 and 1882 I0421 10:54:22.434566 47076539613632 model_training_utils.py:505] Train Step: 1183/2100 / loss = 0.665283203125 I0421 10:54:22.434993 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.67 examples/second between steps 1882 and 1883 I0421 10:54:23.517053 47076539613632 model_training_utils.py:505] Train Step: 1184/2100 / loss = 0.53729248046875 I0421 10:54:23.517476 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 1883 and 1884 I0421 10:54:24.600224 47076539613632 model_training_utils.py:505] Train Step: 1185/2100 / loss = 0.79248046875 I0421 10:54:24.600651 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 1884 and 1885 I0421 10:54:25.683887 47076539613632 model_training_utils.py:505] Train Step: 1186/2100 / loss = 0.916748046875 I0421 10:54:25.684321 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 1885 and 1886 I0421 10:54:26.765638 47076539613632 model_training_utils.py:505] Train Step: 1187/2100 / loss = 0.893798828125 I0421 10:54:26.766067 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 1886 and 1887 I0421 10:54:27.847150 47076539613632 model_training_utils.py:505] Train Step: 1188/2100 / loss = 1.0419921875 I0421 10:54:27.847565 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 1887 and 1888 I0421 10:54:28.928611 47076539613632 model_training_utils.py:505] Train Step: 1189/2100 / loss = 0.844970703125 I0421 10:54:28.929027 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.25 examples/second between steps 1888 and 1889 I0421 10:54:30.010847 47076539613632 model_training_utils.py:505] Train Step: 1190/2100 / loss = 0.8135986328125 I0421 10:54:30.011267 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 1889 and 1890 I0421 10:54:31.095426 47076539613632 model_training_utils.py:505] Train Step: 1191/2100 / loss = 1.102783203125 I0421 10:54:31.095851 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 1890 and 1891 I0421 10:54:32.177403 47076539613632 model_training_utils.py:505] Train Step: 1192/2100 / loss = 1.20556640625 I0421 10:54:32.177825 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 1891 and 1892 I0421 10:54:33.261362 47076539613632 model_training_utils.py:505] Train Step: 1193/2100 / loss = 1.10693359375 I0421 10:54:33.261789 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 1892 and 1893 I0421 10:54:34.342564 47076539613632 model_training_utils.py:505] Train Step: 1194/2100 / loss = 0.998046875 I0421 10:54:34.342998 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 1893 and 1894 I0421 10:54:35.424999 47076539613632 model_training_utils.py:505] Train Step: 1195/2100 / loss = 1.0970458984375 I0421 10:54:35.425435 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 1894 and 1895 I0421 10:54:36.505751 47076539613632 model_training_utils.py:505] Train Step: 1196/2100 / loss = 1.192138671875 I0421 10:54:36.506167 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.31 examples/second between steps 1895 and 1896 I0421 10:54:37.587426 47076539613632 model_training_utils.py:505] Train Step: 1197/2100 / loss = 1.0601806640625 I0421 10:54:37.587843 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 1896 and 1897 I0421 10:54:38.669015 47076539613632 model_training_utils.py:505] Train Step: 1198/2100 / loss = 0.97705078125 I0421 10:54:38.669434 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 1897 and 1898 I0421 10:54:39.749688 47076539613632 model_training_utils.py:505] Train Step: 1199/2100 / loss = 1.251953125 I0421 10:54:39.750105 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.30 examples/second between steps 1898 and 1899 I0421 10:54:40.829445 47076539613632 model_training_utils.py:505] Train Step: 1200/2100 / loss = 1.2822265625 I0421 10:54:40.829872 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.41 examples/second between steps 1899 and 1900 I0421 10:54:41.909152 47076539613632 model_training_utils.py:505] Train Step: 1201/2100 / loss = 1.1689453125 I0421 10:54:41.909586 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.41 examples/second between steps 1900 and 1901 I0421 10:54:42.991694 47076539613632 model_training_utils.py:505] Train Step: 1202/2100 / loss = 1.2255859375 I0421 10:54:42.992125 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 1901 and 1902 I0421 10:54:44.073365 47076539613632 model_training_utils.py:505] Train Step: 1203/2100 / loss = 1.595458984375 I0421 10:54:44.073787 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 1902 and 1903 I0421 10:54:45.156413 47076539613632 model_training_utils.py:505] Train Step: 1204/2100 / loss = 1.45458984375 I0421 10:54:45.156849 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 1903 and 1904 I0421 10:54:46.239885 47076539613632 model_training_utils.py:505] Train Step: 1205/2100 / loss = 1.206787109375 I0421 10:54:46.240316 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 1904 and 1905 I0421 10:54:47.321747 47076539613632 model_training_utils.py:505] Train Step: 1206/2100 / loss = 1.3551025390625 I0421 10:54:47.322175 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 1905 and 1906 I0421 10:54:48.404913 47076539613632 model_training_utils.py:505] Train Step: 1207/2100 / loss = 1.5537109375 I0421 10:54:48.405347 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 1906 and 1907 I0421 10:54:49.485266 47076539613632 model_training_utils.py:505] Train Step: 1208/2100 / loss = 1.156982421875 I0421 10:54:49.485696 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.36 examples/second between steps 1907 and 1908 I0421 10:54:50.569225 47076539613632 model_training_utils.py:505] Train Step: 1209/2100 / loss = 0.7293701171875 I0421 10:54:50.569658 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 1908 and 1909 I0421 10:54:51.652101 47076539613632 model_training_utils.py:505] Train Step: 1210/2100 / loss = 0.901611328125 I0421 10:54:51.652518 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.07 examples/second between steps 1909 and 1910 I0421 10:54:52.735021 47076539613632 model_training_utils.py:505] Train Step: 1211/2100 / loss = 0.8760986328125 I0421 10:54:52.735449 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.07 examples/second between steps 1910 and 1911 I0421 10:54:53.817368 47076539613632 model_training_utils.py:505] Train Step: 1212/2100 / loss = 1.10205078125 I0421 10:54:53.817797 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 1911 and 1912 I0421 10:54:54.899889 47076539613632 model_training_utils.py:505] Train Step: 1213/2100 / loss = 0.9554443359375 I0421 10:54:54.900326 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 1912 and 1913 I0421 10:54:55.982573 47076539613632 model_training_utils.py:505] Train Step: 1214/2100 / loss = 1.166015625 I0421 10:54:55.982996 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 1913 and 1914 I0421 10:54:57.064444 47076539613632 model_training_utils.py:505] Train Step: 1215/2100 / loss = 1.363525390625 I0421 10:54:57.064864 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 1914 and 1915 I0421 10:54:58.145656 47076539613632 model_training_utils.py:505] Train Step: 1216/2100 / loss = 1.0169677734375 I0421 10:54:58.146094 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 1915 and 1916 I0421 10:54:59.226859 47076539613632 model_training_utils.py:505] Train Step: 1217/2100 / loss = 1.0509033203125 I0421 10:54:59.227294 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 1916 and 1917 I0421 10:55:00.307858 47076539613632 model_training_utils.py:505] Train Step: 1218/2100 / loss = 0.947021484375 I0421 10:55:00.308294 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.29 examples/second between steps 1917 and 1918 I0421 10:55:01.390664 47076539613632 model_training_utils.py:505] Train Step: 1219/2100 / loss = 0.939453125 I0421 10:55:01.391088 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 1918 and 1919 I0421 10:55:02.472212 47076539613632 model_training_utils.py:505] Train Step: 1220/2100 / loss = 0.9088134765625 I0421 10:55:02.472647 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 1919 and 1920 I0421 10:55:03.553500 47076539613632 model_training_utils.py:505] Train Step: 1221/2100 / loss = 0.853271484375 I0421 10:55:03.553926 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 1920 and 1921 I0421 10:55:04.636086 47076539613632 model_training_utils.py:505] Train Step: 1222/2100 / loss = 0.80908203125 I0421 10:55:04.636504 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 1921 and 1922 I0421 10:55:05.721218 47076539613632 model_training_utils.py:505] Train Step: 1223/2100 / loss = 0.7862548828125 I0421 10:55:05.721648 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 1922 and 1923 I0421 10:55:06.805697 47076539613632 model_training_utils.py:505] Train Step: 1224/2100 / loss = 0.757568359375 I0421 10:55:06.806127 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 1923 and 1924 I0421 10:55:07.888897 47076539613632 model_training_utils.py:505] Train Step: 1225/2100 / loss = 0.62225341796875 I0421 10:55:07.889332 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 1924 and 1925 I0421 10:55:08.972590 47076539613632 model_training_utils.py:505] Train Step: 1226/2100 / loss = 0.91162109375 I0421 10:55:08.973017 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 1925 and 1926 I0421 10:55:10.059128 47076539613632 model_training_utils.py:505] Train Step: 1227/2100 / loss = 1.09423828125 I0421 10:55:10.059563 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 1926 and 1927 I0421 10:55:11.144008 47076539613632 model_training_utils.py:505] Train Step: 1228/2100 / loss = 1.330322265625 I0421 10:55:11.144443 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 1927 and 1928 I0421 10:55:12.229600 47076539613632 model_training_utils.py:505] Train Step: 1229/2100 / loss = 0.863037109375 I0421 10:55:12.230027 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 1928 and 1929 I0421 10:55:13.313573 47076539613632 model_training_utils.py:505] Train Step: 1230/2100 / loss = 1.050048828125 I0421 10:55:13.314014 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 1929 and 1930 I0421 10:55:14.398669 47076539613632 model_training_utils.py:505] Train Step: 1231/2100 / loss = 1.11376953125 I0421 10:55:14.399102 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 1930 and 1931 I0421 10:55:15.484343 47076539613632 model_training_utils.py:505] Train Step: 1232/2100 / loss = 1.635498046875 I0421 10:55:15.484769 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 1931 and 1932 I0421 10:55:16.570976 47076539613632 model_training_utils.py:505] Train Step: 1233/2100 / loss = 1.245849609375 I0421 10:55:16.571412 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 1932 and 1933 I0421 10:55:17.653395 47076539613632 model_training_utils.py:505] Train Step: 1234/2100 / loss = 0.8402099609375 I0421 10:55:17.653805 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 1933 and 1934 I0421 10:55:18.737642 47076539613632 model_training_utils.py:505] Train Step: 1235/2100 / loss = 0.9281005859375 I0421 10:55:18.738050 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 1934 and 1935 I0421 10:55:19.820512 47076539613632 model_training_utils.py:505] Train Step: 1236/2100 / loss = 1.09375 I0421 10:55:19.820940 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 1935 and 1936 I0421 10:55:20.902325 47076539613632 model_training_utils.py:505] Train Step: 1237/2100 / loss = 1.221923828125 I0421 10:55:20.902749 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 1936 and 1937 I0421 10:55:21.985694 47076539613632 model_training_utils.py:505] Train Step: 1238/2100 / loss = 1.294189453125 I0421 10:55:21.986116 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 1937 and 1938 I0421 10:55:23.067918 47076539613632 model_training_utils.py:505] Train Step: 1239/2100 / loss = 0.861572265625 I0421 10:55:23.068348 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 1938 and 1939 I0421 10:55:24.154022 47076539613632 model_training_utils.py:505] Train Step: 1240/2100 / loss = 0.8148193359375 I0421 10:55:24.154449 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.70 examples/second between steps 1939 and 1940 I0421 10:55:25.237233 47076539613632 model_training_utils.py:505] Train Step: 1241/2100 / loss = 0.8360595703125 I0421 10:55:25.237664 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 1940 and 1941 I0421 10:55:26.319400 47076539613632 model_training_utils.py:505] Train Step: 1242/2100 / loss = 0.989013671875 I0421 10:55:26.319831 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 1941 and 1942 I0421 10:55:27.402662 47076539613632 model_training_utils.py:505] Train Step: 1243/2100 / loss = 1.143798828125 I0421 10:55:27.403085 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 1942 and 1943 I0421 10:55:28.483689 47076539613632 model_training_utils.py:505] Train Step: 1244/2100 / loss = 0.961181640625 I0421 10:55:28.484112 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.27 examples/second between steps 1943 and 1944 I0421 10:55:29.565944 47076539613632 model_training_utils.py:505] Train Step: 1245/2100 / loss = 1.0849609375 I0421 10:55:29.566380 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 1944 and 1945 I0421 10:55:30.648339 47076539613632 model_training_utils.py:505] Train Step: 1246/2100 / loss = 1.46533203125 I0421 10:55:30.648750 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 1945 and 1946 I0421 10:55:31.730499 47076539613632 model_training_utils.py:505] Train Step: 1247/2100 / loss = 1.3916015625 I0421 10:55:31.730921 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 1946 and 1947 I0421 10:55:32.813877 47076539613632 model_training_utils.py:505] Train Step: 1248/2100 / loss = 1.230712890625 I0421 10:55:32.814307 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 1947 and 1948 I0421 10:55:33.898104 47076539613632 model_training_utils.py:505] Train Step: 1249/2100 / loss = 1.095458984375 I0421 10:55:33.898530 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 1948 and 1949 I0421 10:55:34.978644 47076539613632 model_training_utils.py:505] Train Step: 1250/2100 / loss = 1.1038818359375 I0421 10:55:34.979072 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.30 examples/second between steps 1949 and 1950 I0421 10:55:36.064042 47076539613632 model_training_utils.py:505] Train Step: 1251/2100 / loss = 1.22900390625 I0421 10:55:36.064474 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 1950 and 1951 I0421 10:55:37.149436 47076539613632 model_training_utils.py:505] Train Step: 1252/2100 / loss = 1.0675048828125 I0421 10:55:37.149819 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 1951 and 1952 I0421 10:55:38.233994 47076539613632 model_training_utils.py:505] Train Step: 1253/2100 / loss = 0.8753662109375 I0421 10:55:38.234383 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 1952 and 1953 I0421 10:55:39.318011 47076539613632 model_training_utils.py:505] Train Step: 1254/2100 / loss = 1.155029296875 I0421 10:55:39.318405 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 1953 and 1954 I0421 10:55:40.398483 47076539613632 model_training_utils.py:505] Train Step: 1255/2100 / loss = 1.1611328125 I0421 10:55:40.398866 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.36 examples/second between steps 1954 and 1955 I0421 10:55:41.482731 47076539613632 model_training_utils.py:505] Train Step: 1256/2100 / loss = 1.1182861328125 I0421 10:55:41.483114 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 1955 and 1956 I0421 10:55:42.565292 47076539613632 model_training_utils.py:505] Train Step: 1257/2100 / loss = 1.084716796875 I0421 10:55:42.565674 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 1956 and 1957 I0421 10:55:43.643664 47076539613632 model_training_utils.py:505] Train Step: 1258/2100 / loss = 1.0367431640625 I0421 10:55:43.644046 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.59 examples/second between steps 1957 and 1958 I0421 10:55:44.729025 47076539613632 model_training_utils.py:505] Train Step: 1259/2100 / loss = 1.203857421875 I0421 10:55:44.729414 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 1958 and 1959 I0421 10:55:45.814631 47076539613632 model_training_utils.py:505] Train Step: 1260/2100 / loss = 0.958251953125 I0421 10:55:45.815016 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 1959 and 1960 I0421 10:55:46.900588 47076539613632 model_training_utils.py:505] Train Step: 1261/2100 / loss = 1.882568359375 I0421 10:55:46.900967 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 1960 and 1961 I0421 10:55:47.984512 47076539613632 model_training_utils.py:505] Train Step: 1262/2100 / loss = 3.03759765625 I0421 10:55:47.984894 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 1961 and 1962 I0421 10:55:49.067658 47076539613632 model_training_utils.py:505] Train Step: 1263/2100 / loss = 3.06982421875 I0421 10:55:49.068045 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 1962 and 1963 I0421 10:55:50.150640 47076539613632 model_training_utils.py:505] Train Step: 1264/2100 / loss = 1.87109375 I0421 10:55:50.151024 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 1963 and 1964 I0421 10:55:51.237840 47076539613632 model_training_utils.py:505] Train Step: 1265/2100 / loss = 1.940673828125 I0421 10:55:51.238227 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 1964 and 1965 I0421 10:55:52.322836 47076539613632 model_training_utils.py:505] Train Step: 1266/2100 / loss = 1.955322265625 I0421 10:55:52.323219 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 1965 and 1966 I0421 10:55:53.404562 47076539613632 model_training_utils.py:505] Train Step: 1267/2100 / loss = 1.096923828125 I0421 10:55:53.404949 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.25 examples/second between steps 1966 and 1967 I0421 10:55:54.487729 47076539613632 model_training_utils.py:505] Train Step: 1268/2100 / loss = 0.642822265625 I0421 10:55:54.488113 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 1967 and 1968 I0421 10:55:55.570487 47076539613632 model_training_utils.py:505] Train Step: 1269/2100 / loss = 0.848876953125 I0421 10:55:55.570873 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 1968 and 1969 I0421 10:55:56.655552 47076539613632 model_training_utils.py:505] Train Step: 1270/2100 / loss = 0.806396484375 I0421 10:55:56.655930 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 1969 and 1970 I0421 10:55:57.742065 47076539613632 model_training_utils.py:505] Train Step: 1271/2100 / loss = 1.0511474609375 I0421 10:55:57.742457 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 1970 and 1971 I0421 10:55:58.826685 47076539613632 model_training_utils.py:505] Train Step: 1272/2100 / loss = 0.986572265625 I0421 10:55:58.827081 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 1971 and 1972 I0421 10:55:59.911682 47076539613632 model_training_utils.py:505] Train Step: 1273/2100 / loss = 1.072021484375 I0421 10:55:59.912061 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 1972 and 1973 I0421 10:56:00.998037 47076539613632 model_training_utils.py:505] Train Step: 1274/2100 / loss = 1.51513671875 I0421 10:56:00.998426 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 1973 and 1974 I0421 10:56:02.088098 47076539613632 model_training_utils.py:505] Train Step: 1275/2100 / loss = 1.25830078125 I0421 10:56:02.088487 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.35 examples/second between steps 1974 and 1975 I0421 10:56:03.173796 47076539613632 model_training_utils.py:505] Train Step: 1276/2100 / loss = 1.22265625 I0421 10:56:03.174181 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 1975 and 1976 I0421 10:56:04.259887 47076539613632 model_training_utils.py:505] Train Step: 1277/2100 / loss = 1.744873046875 I0421 10:56:04.260288 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 1976 and 1977 I0421 10:56:05.348261 47076539613632 model_training_utils.py:505] Train Step: 1278/2100 / loss = 2.4189453125 I0421 10:56:05.348658 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.54 examples/second between steps 1977 and 1978 I0421 10:56:06.439598 47076539613632 model_training_utils.py:505] Train Step: 1279/2100 / loss = 1.295166015625 I0421 10:56:06.439973 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.23 examples/second between steps 1978 and 1979 I0421 10:56:07.524160 47076539613632 model_training_utils.py:505] Train Step: 1280/2100 / loss = 1.11376953125 I0421 10:56:07.524548 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 1979 and 1980 I0421 10:56:08.610114 47076539613632 model_training_utils.py:505] Train Step: 1281/2100 / loss = 0.952392578125 I0421 10:56:08.610503 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 1980 and 1981 I0421 10:56:09.695623 47076539613632 model_training_utils.py:505] Train Step: 1282/2100 / loss = 0.9786376953125 I0421 10:56:09.696002 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 1981 and 1982 I0421 10:56:10.779821 47076539613632 model_training_utils.py:505] Train Step: 1283/2100 / loss = 0.893798828125 I0421 10:56:10.780200 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 1982 and 1983 I0421 10:56:11.863312 47076539613632 model_training_utils.py:505] Train Step: 1284/2100 / loss = 1.0372314453125 I0421 10:56:11.863697 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 1983 and 1984 I0421 10:56:12.949637 47076539613632 model_training_utils.py:505] Train Step: 1285/2100 / loss = 1.19677734375 I0421 10:56:12.950022 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 1984 and 1985 I0421 10:56:14.037544 47076539613632 model_training_utils.py:505] Train Step: 1286/2100 / loss = 1.47705078125 I0421 10:56:14.037933 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.59 examples/second between steps 1985 and 1986 I0421 10:56:15.122255 47076539613632 model_training_utils.py:505] Train Step: 1287/2100 / loss = 1.484375 I0421 10:56:15.122637 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 1986 and 1987 I0421 10:56:16.204763 47076539613632 model_training_utils.py:505] Train Step: 1288/2100 / loss = 1.101806640625 I0421 10:56:16.205140 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 1987 and 1988 I0421 10:56:17.287631 47076539613632 model_training_utils.py:505] Train Step: 1289/2100 / loss = 1.1097412109375 I0421 10:56:17.288010 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 1988 and 1989 I0421 10:56:18.372932 47076539613632 model_training_utils.py:505] Train Step: 1290/2100 / loss = 1.7059326171875 I0421 10:56:18.373319 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 1989 and 1990 I0421 10:56:19.454456 47076539613632 model_training_utils.py:505] Train Step: 1291/2100 / loss = 2.720703125 I0421 10:56:19.454838 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.27 examples/second between steps 1990 and 1991 I0421 10:56:20.537250 47076539613632 model_training_utils.py:505] Train Step: 1292/2100 / loss = 2.041259765625 I0421 10:56:20.537641 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 1991 and 1992 I0421 10:56:21.622564 47076539613632 model_training_utils.py:505] Train Step: 1293/2100 / loss = 1.54541015625 I0421 10:56:21.622956 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 1992 and 1993 I0421 10:56:22.708037 47076539613632 model_training_utils.py:505] Train Step: 1294/2100 / loss = 1.308349609375 I0421 10:56:22.708419 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 1993 and 1994 I0421 10:56:23.794882 47076539613632 model_training_utils.py:505] Train Step: 1295/2100 / loss = 0.8817138671875 I0421 10:56:23.795265 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 1994 and 1995 I0421 10:56:24.879812 47076539613632 model_training_utils.py:505] Train Step: 1296/2100 / loss = 1.15625 I0421 10:56:24.880192 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 1995 and 1996 I0421 10:56:25.967364 47076539613632 model_training_utils.py:505] Train Step: 1297/2100 / loss = 1.2003173828125 I0421 10:56:25.967744 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.64 examples/second between steps 1996 and 1997 I0421 10:56:27.051506 47076539613632 model_training_utils.py:505] Train Step: 1298/2100 / loss = 1.375244140625 I0421 10:56:27.051900 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 1997 and 1998 I0421 10:56:28.136239 47076539613632 model_training_utils.py:505] Train Step: 1299/2100 / loss = 1.28369140625 I0421 10:56:28.136631 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 1998 and 1999 I0421 10:56:29.224359 47076539613632 model_training_utils.py:505] Train Step: 1300/2100 / loss = 1.5322265625 I0421 10:56:29.224737 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.54 examples/second between steps 1999 and 2000 I0421 10:56:30.314162 47076539613632 model_training_utils.py:505] Train Step: 1301/2100 / loss = 1.19140625 I0421 10:56:30.314547 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.36 examples/second between steps 2000 and 2001 I0421 10:56:31.403783 47076539613632 model_training_utils.py:505] Train Step: 1302/2100 / loss = 1.87109375 I0421 10:56:31.404166 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.39 examples/second between steps 2001 and 2002 I0421 10:56:32.494476 47076539613632 model_training_utils.py:505] Train Step: 1303/2100 / loss = 1.76220703125 I0421 10:56:32.494859 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.27 examples/second between steps 2002 and 2003 I0421 10:56:33.580059 47076539613632 model_training_utils.py:505] Train Step: 1304/2100 / loss = 1.466064453125 I0421 10:56:33.580445 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 2003 and 2004 I0421 10:56:34.665001 47076539613632 model_training_utils.py:505] Train Step: 1305/2100 / loss = 0.9449462890625 I0421 10:56:34.665396 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 2004 and 2005 I0421 10:56:35.742934 47076539613632 model_training_utils.py:505] Train Step: 1306/2100 / loss = 0.9364013671875 I0421 10:56:35.743326 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.62 examples/second between steps 2005 and 2006 I0421 10:56:36.825926 47076539613632 model_training_utils.py:505] Train Step: 1307/2100 / loss = 0.9718017578125 I0421 10:56:36.826321 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 2006 and 2007 I0421 10:56:37.912619 47076539613632 model_training_utils.py:505] Train Step: 1308/2100 / loss = 0.9111328125 I0421 10:56:37.913000 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 2007 and 2008 I0421 10:56:38.997347 47076539613632 model_training_utils.py:505] Train Step: 1309/2100 / loss = 0.930419921875 I0421 10:56:38.997737 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 2008 and 2009 I0421 10:56:40.082257 47076539613632 model_training_utils.py:505] Train Step: 1310/2100 / loss = 0.91259765625 I0421 10:56:40.082642 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 2009 and 2010 I0421 10:56:41.167914 47076539613632 model_training_utils.py:505] Train Step: 1311/2100 / loss = 1.151123046875 I0421 10:56:41.168306 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 2010 and 2011 I0421 10:56:42.252167 47076539613632 model_training_utils.py:505] Train Step: 1312/2100 / loss = 1.2333984375 I0421 10:56:42.252560 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 2011 and 2012 I0421 10:56:43.335234 47076539613632 model_training_utils.py:505] Train Step: 1313/2100 / loss = 0.8731689453125 I0421 10:56:43.335622 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 2012 and 2013 I0421 10:56:44.418907 47076539613632 model_training_utils.py:505] Train Step: 1314/2100 / loss = 1.211181640625 I0421 10:56:44.419317 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 2013 and 2014 I0421 10:56:45.500782 47076539613632 model_training_utils.py:505] Train Step: 1315/2100 / loss = 0.7628173828125 I0421 10:56:45.501163 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 2014 and 2015 I0421 10:56:46.586881 47076539613632 model_training_utils.py:505] Train Step: 1316/2100 / loss = 0.9052734375 I0421 10:56:46.587263 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 2015 and 2016 I0421 10:56:47.670436 47076539613632 model_training_utils.py:505] Train Step: 1317/2100 / loss = 1.19189453125 I0421 10:56:47.670822 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 2016 and 2017 I0421 10:56:48.753949 47076539613632 model_training_utils.py:505] Train Step: 1318/2100 / loss = 1.541259765625 I0421 10:56:48.754342 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 2017 and 2018 I0421 10:56:49.841529 47076539613632 model_training_utils.py:505] Train Step: 1319/2100 / loss = 1.3349609375 I0421 10:56:49.841912 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.56 examples/second between steps 2018 and 2019 I0421 10:56:50.931097 47076539613632 model_training_utils.py:505] Train Step: 1320/2100 / loss = 1.997802734375 I0421 10:56:50.931529 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.34 examples/second between steps 2019 and 2020 I0421 10:56:52.018348 47076539613632 model_training_utils.py:505] Train Step: 1321/2100 / loss = 2.2763671875 I0421 10:56:52.018784 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.62 examples/second between steps 2020 and 2021 I0421 10:56:53.103773 47076539613632 model_training_utils.py:505] Train Step: 1322/2100 / loss = 2.4814453125 I0421 10:56:53.104202 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 2021 and 2022 I0421 10:56:54.189960 47076539613632 model_training_utils.py:505] Train Step: 1323/2100 / loss = 1.77783203125 I0421 10:56:54.190403 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 2022 and 2023 I0421 10:56:55.272007 47076539613632 model_training_utils.py:505] Train Step: 1324/2100 / loss = 1.2315673828125 I0421 10:56:55.272440 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 2023 and 2024 I0421 10:56:56.357180 47076539613632 model_training_utils.py:505] Train Step: 1325/2100 / loss = 1.1649169921875 I0421 10:56:56.357619 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 2024 and 2025 I0421 10:56:57.444439 47076539613632 model_training_utils.py:505] Train Step: 1326/2100 / loss = 1.09033203125 I0421 10:56:57.444864 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.63 examples/second between steps 2025 and 2026 I0421 10:56:58.528932 47076539613632 model_training_utils.py:505] Train Step: 1327/2100 / loss = 0.799072265625 I0421 10:56:58.529372 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 2026 and 2027 I0421 10:56:59.612361 47076539613632 model_training_utils.py:505] Train Step: 1328/2100 / loss = 0.670166015625 I0421 10:56:59.612793 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 2027 and 2028 I0421 10:57:00.697109 47076539613632 model_training_utils.py:505] Train Step: 1329/2100 / loss = 0.716064453125 I0421 10:57:00.697555 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 2028 and 2029 I0421 10:57:01.779134 47076539613632 model_training_utils.py:505] Train Step: 1330/2100 / loss = 1.10546875 I0421 10:57:01.779567 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 2029 and 2030 I0421 10:57:02.867947 47076539613632 model_training_utils.py:505] Train Step: 1331/2100 / loss = 0.796630859375 I0421 10:57:02.868337 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.46 examples/second between steps 2030 and 2031 I0421 10:57:03.952734 47076539613632 model_training_utils.py:505] Train Step: 1332/2100 / loss = 1.108642578125 I0421 10:57:03.953114 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 2031 and 2032 I0421 10:57:05.035764 47076539613632 model_training_utils.py:505] Train Step: 1333/2100 / loss = 0.855712890625 I0421 10:57:05.036151 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 2032 and 2033 I0421 10:57:06.119376 47076539613632 model_training_utils.py:505] Train Step: 1334/2100 / loss = 0.9703369140625 I0421 10:57:06.119759 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 2033 and 2034 I0421 10:57:07.202216 47076539613632 model_training_utils.py:505] Train Step: 1335/2100 / loss = 0.9932861328125 I0421 10:57:07.202612 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 2034 and 2035 I0421 10:57:08.284937 47076539613632 model_training_utils.py:505] Train Step: 1336/2100 / loss = 0.8895263671875 I0421 10:57:08.285334 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 2035 and 2036 I0421 10:57:09.364993 47076539613632 model_training_utils.py:505] Train Step: 1337/2100 / loss = 1.155029296875 I0421 10:57:09.365385 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.37 examples/second between steps 2036 and 2037 I0421 10:57:10.449580 47076539613632 model_training_utils.py:505] Train Step: 1338/2100 / loss = 0.92333984375 I0421 10:57:10.449980 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 2037 and 2038 I0421 10:57:11.532649 47076539613632 model_training_utils.py:505] Train Step: 1339/2100 / loss = 0.8663330078125 I0421 10:57:11.533030 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 2038 and 2039 I0421 10:57:12.613939 47076539613632 model_training_utils.py:505] Train Step: 1340/2100 / loss = 0.9495849609375 I0421 10:57:12.614327 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 2039 and 2040 I0421 10:57:13.697698 47076539613632 model_training_utils.py:505] Train Step: 1341/2100 / loss = 1.197509765625 I0421 10:57:13.698081 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 2040 and 2041 I0421 10:57:14.780497 47076539613632 model_training_utils.py:505] Train Step: 1342/2100 / loss = 1.089599609375 I0421 10:57:14.780925 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 2041 and 2042 I0421 10:57:15.863250 47076539613632 model_training_utils.py:505] Train Step: 1343/2100 / loss = 1.0797119140625 I0421 10:57:15.863677 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 2042 and 2043 I0421 10:57:16.945248 47076539613632 model_training_utils.py:505] Train Step: 1344/2100 / loss = 0.8990478515625 I0421 10:57:16.945677 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 2043 and 2044 I0421 10:57:18.027911 47076539613632 model_training_utils.py:505] Train Step: 1345/2100 / loss = 0.83056640625 I0421 10:57:18.028347 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 2044 and 2045 I0421 10:57:19.109086 47076539613632 model_training_utils.py:505] Train Step: 1346/2100 / loss = 1.05712890625 I0421 10:57:19.109525 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 2045 and 2046 I0421 10:57:20.190124 47076539613632 model_training_utils.py:505] Train Step: 1347/2100 / loss = 1.041748046875 I0421 10:57:20.190552 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.32 examples/second between steps 2046 and 2047 I0421 10:57:21.272483 47076539613632 model_training_utils.py:505] Train Step: 1348/2100 / loss = 0.9569091796875 I0421 10:57:21.272905 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 2047 and 2048 I0421 10:57:22.353709 47076539613632 model_training_utils.py:505] Train Step: 1349/2100 / loss = 1.236328125 I0421 10:57:22.354147 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 2048 and 2049 I0421 10:57:23.436195 47076539613632 model_training_utils.py:505] Train Step: 1350/2100 / loss = 1.311767578125 I0421 10:57:23.436624 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 2049 and 2050 I0421 10:57:24.514302 47076539613632 model_training_utils.py:505] Train Step: 1351/2100 / loss = 2.12548828125 I0421 10:57:24.514731 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.60 examples/second between steps 2050 and 2051 I0421 10:57:25.596101 47076539613632 model_training_utils.py:505] Train Step: 1352/2100 / loss = 1.504638671875 I0421 10:57:25.596530 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 2051 and 2052 I0421 10:57:26.675919 47076539613632 model_training_utils.py:505] Train Step: 1353/2100 / loss = 1.054931640625 I0421 10:57:26.676345 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.43 examples/second between steps 2052 and 2053 I0421 10:57:27.758358 47076539613632 model_training_utils.py:505] Train Step: 1354/2100 / loss = 1.1495361328125 I0421 10:57:27.758761 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 2053 and 2054 I0421 10:57:28.843572 47076539613632 model_training_utils.py:505] Train Step: 1355/2100 / loss = 1.220458984375 I0421 10:57:28.843997 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 2054 and 2055 I0421 10:57:29.924918 47076539613632 model_training_utils.py:505] Train Step: 1356/2100 / loss = 1.2099609375 I0421 10:57:29.925351 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 2055 and 2056 I0421 10:57:31.008743 47076539613632 model_training_utils.py:505] Train Step: 1357/2100 / loss = 1.3275146484375 I0421 10:57:31.009162 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 2056 and 2057 I0421 10:57:32.090492 47076539613632 model_training_utils.py:505] Train Step: 1358/2100 / loss = 0.7791748046875 I0421 10:57:32.090912 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 2057 and 2058 I0421 10:57:33.173777 47076539613632 model_training_utils.py:505] Train Step: 1359/2100 / loss = 0.97021484375 I0421 10:57:33.174209 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 2058 and 2059 I0421 10:57:34.256091 47076539613632 model_training_utils.py:505] Train Step: 1360/2100 / loss = 1.25537109375 I0421 10:57:34.256522 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 2059 and 2060 I0421 10:57:35.337001 47076539613632 model_training_utils.py:505] Train Step: 1361/2100 / loss = 1.3173828125 I0421 10:57:35.337441 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.29 examples/second between steps 2060 and 2061 I0421 10:57:36.421427 47076539613632 model_training_utils.py:505] Train Step: 1362/2100 / loss = 1.115478515625 I0421 10:57:36.421852 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 2061 and 2062 I0421 10:57:37.505270 47076539613632 model_training_utils.py:505] Train Step: 1363/2100 / loss = 0.899169921875 I0421 10:57:37.505704 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 2062 and 2063 I0421 10:57:38.590958 47076539613632 model_training_utils.py:505] Train Step: 1364/2100 / loss = 0.917236328125 I0421 10:57:38.591390 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 2063 and 2064 I0421 10:57:39.674859 47076539613632 model_training_utils.py:505] Train Step: 1365/2100 / loss = 0.8585205078125 I0421 10:57:39.675266 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 2064 and 2065 I0421 10:57:40.760286 47076539613632 model_training_utils.py:505] Train Step: 1366/2100 / loss = 0.9852294921875 I0421 10:57:40.760727 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 2065 and 2066 I0421 10:57:41.845504 47076539613632 model_training_utils.py:505] Train Step: 1367/2100 / loss = 0.974853515625 I0421 10:57:41.845909 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 2066 and 2067 I0421 10:57:42.923945 47076539613632 model_training_utils.py:505] Train Step: 1368/2100 / loss = 0.984619140625 I0421 10:57:42.924372 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.60 examples/second between steps 2067 and 2068 I0421 10:57:44.009422 47076539613632 model_training_utils.py:505] Train Step: 1369/2100 / loss = 1.276611328125 I0421 10:57:44.009824 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 2068 and 2069 I0421 10:57:45.092396 47076539613632 model_training_utils.py:505] Train Step: 1370/2100 / loss = 1.276123046875 I0421 10:57:45.092818 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 2069 and 2070 I0421 10:57:46.175848 47076539613632 model_training_utils.py:505] Train Step: 1371/2100 / loss = 1.339111328125 I0421 10:57:46.176266 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 2070 and 2071 I0421 10:57:47.261762 47076539613632 model_training_utils.py:505] Train Step: 1372/2100 / loss = 1.541015625 I0421 10:57:47.262186 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 2071 and 2072 I0421 10:57:48.344805 47076539613632 model_training_utils.py:505] Train Step: 1373/2100 / loss = 1.137451171875 I0421 10:57:48.345230 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 2072 and 2073 I0421 10:57:49.428016 47076539613632 model_training_utils.py:505] Train Step: 1374/2100 / loss = 0.992431640625 I0421 10:57:49.428467 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 2073 and 2074 I0421 10:57:50.513329 47076539613632 model_training_utils.py:505] Train Step: 1375/2100 / loss = 0.796142578125 I0421 10:57:50.513750 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 2074 and 2075 I0421 10:57:51.595134 47076539613632 model_training_utils.py:505] Train Step: 1376/2100 / loss = 0.7479248046875 I0421 10:57:51.595559 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 2075 and 2076 I0421 10:57:52.675224 47076539613632 model_training_utils.py:505] Train Step: 1377/2100 / loss = 0.8878173828125 I0421 10:57:52.675654 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.39 examples/second between steps 2076 and 2077 I0421 10:57:53.759373 47076539613632 model_training_utils.py:505] Train Step: 1378/2100 / loss = 0.83154296875 I0421 10:57:53.759796 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 2077 and 2078 I0421 10:57:54.841131 47076539613632 model_training_utils.py:505] Train Step: 1379/2100 / loss = 0.8544921875 I0421 10:57:54.841568 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 2078 and 2079 I0421 10:57:55.925749 47076539613632 model_training_utils.py:505] Train Step: 1380/2100 / loss = 0.7730712890625 I0421 10:57:55.926179 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 2079 and 2080 I0421 10:57:57.009552 47076539613632 model_training_utils.py:505] Train Step: 1381/2100 / loss = 1.11279296875 I0421 10:57:57.009974 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 2080 and 2081 I0421 10:57:58.088661 47076539613632 model_training_utils.py:505] Train Step: 1382/2100 / loss = 1.22900390625 I0421 10:57:58.089071 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.52 examples/second between steps 2081 and 2082 I0421 10:57:59.170057 47076539613632 model_training_utils.py:505] Train Step: 1383/2100 / loss = 0.9561767578125 I0421 10:57:59.170495 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 2082 and 2083 I0421 10:58:00.250631 47076539613632 model_training_utils.py:505] Train Step: 1384/2100 / loss = 0.8328857421875 I0421 10:58:00.251050 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.36 examples/second between steps 2083 and 2084 I0421 10:58:01.329919 47076539613632 model_training_utils.py:505] Train Step: 1385/2100 / loss = 0.7149658203125 I0421 10:58:01.330351 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.48 examples/second between steps 2084 and 2085 I0421 10:58:02.415207 47076539613632 model_training_utils.py:505] Train Step: 1386/2100 / loss = 0.831298828125 I0421 10:58:02.415638 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 2085 and 2086 I0421 10:58:03.501785 47076539613632 model_training_utils.py:505] Train Step: 1387/2100 / loss = 0.867431640625 I0421 10:58:03.502206 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.70 examples/second between steps 2086 and 2087 I0421 10:58:04.587347 47076539613632 model_training_utils.py:505] Train Step: 1388/2100 / loss = 0.9140625 I0421 10:58:04.587765 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 2087 and 2088 I0421 10:58:05.672329 47076539613632 model_training_utils.py:505] Train Step: 1389/2100 / loss = 1.145751953125 I0421 10:58:05.672739 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 2088 and 2089 I0421 10:58:06.754335 47076539613632 model_training_utils.py:505] Train Step: 1390/2100 / loss = 1.31494140625 I0421 10:58:06.754758 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 2089 and 2090 I0421 10:58:07.833028 47076539613632 model_training_utils.py:505] Train Step: 1391/2100 / loss = 1.510009765625 I0421 10:58:07.833464 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.57 examples/second between steps 2090 and 2091 I0421 10:58:08.913906 47076539613632 model_training_utils.py:505] Train Step: 1392/2100 / loss = 0.976806640625 I0421 10:58:08.914338 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.32 examples/second between steps 2091 and 2092 I0421 10:58:09.998604 47076539613632 model_training_utils.py:505] Train Step: 1393/2100 / loss = 1.142822265625 I0421 10:58:09.999046 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 2092 and 2093 I0421 10:58:11.080231 47076539613632 model_training_utils.py:505] Train Step: 1394/2100 / loss = 1.144775390625 I0421 10:58:11.080668 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 2093 and 2094 I0421 10:58:12.163222 47076539613632 model_training_utils.py:505] Train Step: 1395/2100 / loss = 1.0145263671875 I0421 10:58:12.163659 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 2094 and 2095 I0421 10:58:13.241735 47076539613632 model_training_utils.py:505] Train Step: 1396/2100 / loss = 0.9039306640625 I0421 10:58:13.242152 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.58 examples/second between steps 2095 and 2096 I0421 10:58:14.321872 47076539613632 model_training_utils.py:505] Train Step: 1397/2100 / loss = 0.6424560546875 I0421 10:58:14.322315 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.40 examples/second between steps 2096 and 2097 I0421 10:58:15.397186 47076539613632 model_training_utils.py:505] Train Step: 1398/2100 / loss = 0.836181640625 I0421 10:58:15.397629 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.95 examples/second between steps 2097 and 2098 I0421 10:58:16.479586 47076539613632 model_training_utils.py:505] Train Step: 1399/2100 / loss = 0.909912109375 I0421 10:58:16.480013 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 2098 and 2099 I0421 10:58:17.560334 47076539613632 model_training_utils.py:505] Train Step: 1400/2100 / loss = 0.7618408203125 I0421 10:58:26.056051 47076539613632 model_training_utils.py:49] Saving model as TF checkpoint: /public/home/xuanbaby/DL-TensorFlow/models_r2.3.0/official/nlp/bert/model_squad_v2/ctl_step_1400.ckpt-2 I0421 10:58:26.056570 47076539613632 keras_utils.py:133] TimeHistory: 9.57 seconds, 13.38 examples/second between steps 2099 and 2100 I0421 10:58:27.937270 47076539613632 model_training_utils.py:505] Train Step: 1401/2100 / loss = 0.9248046875 I0421 10:58:27.937665 47076539613632 keras_utils.py:133] TimeHistory: 1.32 seconds, 135914.65 examples/second between steps 2100 and 3501 I0421 10:58:29.022498 47076539613632 model_training_utils.py:505] Train Step: 1402/2100 / loss = 0.59039306640625 I0421 10:58:29.022932 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 3501 and 3502 I0421 10:58:30.105153 47076539613632 model_training_utils.py:505] Train Step: 1403/2100 / loss = 0.85986328125 I0421 10:58:30.105587 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 3502 and 3503 I0421 10:58:31.188868 47076539613632 model_training_utils.py:505] Train Step: 1404/2100 / loss = 1.323974609375 I0421 10:58:31.189297 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 3503 and 3504 I0421 10:58:32.269772 47076539613632 model_training_utils.py:505] Train Step: 1405/2100 / loss = 1.2178955078125 I0421 10:58:32.270190 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.29 examples/second between steps 3504 and 3505 I0421 10:58:33.350827 47076539613632 model_training_utils.py:505] Train Step: 1406/2100 / loss = 1.34326171875 I0421 10:58:33.351245 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.27 examples/second between steps 3505 and 3506 I0421 10:58:34.432333 47076539613632 model_training_utils.py:505] Train Step: 1407/2100 / loss = 1.20849609375 I0421 10:58:34.432768 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 3506 and 3507 I0421 10:58:35.514523 47076539613632 model_training_utils.py:505] Train Step: 1408/2100 / loss = 1.13037109375 I0421 10:58:35.514949 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 3507 and 3508 I0421 10:58:36.597331 47076539613632 model_training_utils.py:505] Train Step: 1409/2100 / loss = 0.9967041015625 I0421 10:58:36.597743 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.07 examples/second between steps 3508 and 3509 I0421 10:58:37.678773 47076539613632 model_training_utils.py:505] Train Step: 1410/2100 / loss = 0.6202392578125 I0421 10:58:37.679190 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 3509 and 3510 I0421 10:58:38.762801 47076539613632 model_training_utils.py:505] Train Step: 1411/2100 / loss = 1.21533203125 I0421 10:58:38.763220 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 3510 and 3511 I0421 10:58:39.845317 47076539613632 model_training_utils.py:505] Train Step: 1412/2100 / loss = 1.14208984375 I0421 10:58:39.845741 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 3511 and 3512 I0421 10:58:40.927002 47076539613632 model_training_utils.py:505] Train Step: 1413/2100 / loss = 0.8612060546875 I0421 10:58:40.927433 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 3512 and 3513 I0421 10:58:42.009942 47076539613632 model_training_utils.py:505] Train Step: 1414/2100 / loss = 1.00146484375 I0421 10:58:42.010391 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 3513 and 3514 I0421 10:58:43.088415 47076539613632 model_training_utils.py:505] Train Step: 1415/2100 / loss = 0.994140625 I0421 10:58:43.088825 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.61 examples/second between steps 3514 and 3515 I0421 10:58:44.168082 47076539613632 model_training_utils.py:505] Train Step: 1416/2100 / loss = 0.86962890625 I0421 10:58:44.168516 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.40 examples/second between steps 3515 and 3516 I0421 10:58:45.249573 47076539613632 model_training_utils.py:505] Train Step: 1417/2100 / loss = 1.1710205078125 I0421 10:58:45.249990 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 3516 and 3517 I0421 10:58:46.333149 47076539613632 model_training_utils.py:505] Train Step: 1418/2100 / loss = 0.7786865234375 I0421 10:58:46.333577 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 3517 and 3518 I0421 10:58:47.414906 47076539613632 model_training_utils.py:505] Train Step: 1419/2100 / loss = 1.159912109375 I0421 10:58:47.415336 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 3518 and 3519 I0421 10:58:48.499354 47076539613632 model_training_utils.py:505] Train Step: 1420/2100 / loss = 1.111572265625 I0421 10:58:48.499782 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 3519 and 3520 I0421 10:58:49.583329 47076539613632 model_training_utils.py:505] Train Step: 1421/2100 / loss = 1.200927734375 I0421 10:58:49.583742 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 3520 and 3521 I0421 10:58:50.668576 47076539613632 model_training_utils.py:505] Train Step: 1422/2100 / loss = 1.25439453125 I0421 10:58:50.668991 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 3521 and 3522 I0421 10:58:51.763358 47076539613632 model_training_utils.py:505] Train Step: 1423/2100 / loss = 1.133056640625 I0421 10:58:51.763778 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.77 examples/second between steps 3522 and 3523 I0421 10:58:52.845076 47076539613632 model_training_utils.py:505] Train Step: 1424/2100 / loss = 1.1007080078125 I0421 10:58:52.845502 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 3523 and 3524 I0421 10:58:53.929359 47076539613632 model_training_utils.py:505] Train Step: 1425/2100 / loss = 1.083984375 I0421 10:58:53.929779 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 3524 and 3525 I0421 10:58:55.010085 47076539613632 model_training_utils.py:505] Train Step: 1426/2100 / loss = 1.097900390625 I0421 10:58:55.010519 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.31 examples/second between steps 3525 and 3526 I0421 10:58:56.095723 47076539613632 model_training_utils.py:505] Train Step: 1427/2100 / loss = 1.0313720703125 I0421 10:58:56.096132 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 3526 and 3527 I0421 10:58:57.179080 47076539613632 model_training_utils.py:505] Train Step: 1428/2100 / loss = 1.0867919921875 I0421 10:58:57.179510 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 3527 and 3528 I0421 10:58:58.271019 47076539613632 model_training_utils.py:505] Train Step: 1429/2100 / loss = 0.9844970703125 I0421 10:58:58.271437 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.10 examples/second between steps 3528 and 3529 I0421 10:58:59.363301 47076539613632 model_training_utils.py:505] Train Step: 1430/2100 / loss = 1.030029296875 I0421 10:58:59.363729 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.07 examples/second between steps 3529 and 3530 I0421 10:59:00.449637 47076539613632 model_training_utils.py:505] Train Step: 1431/2100 / loss = 1.022705078125 I0421 10:59:00.450053 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 3530 and 3531 I0421 10:59:01.539561 47076539613632 model_training_utils.py:505] Train Step: 1432/2100 / loss = 0.8087158203125 I0421 10:59:01.539985 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.28 examples/second between steps 3531 and 3532 I0421 10:59:02.621652 47076539613632 model_training_utils.py:505] Train Step: 1433/2100 / loss = 0.79296875 I0421 10:59:02.622080 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 3532 and 3533 I0421 10:59:03.708516 47076539613632 model_training_utils.py:505] Train Step: 1434/2100 / loss = 0.7435302734375 I0421 10:59:03.708944 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.61 examples/second between steps 3533 and 3534 I0421 10:59:04.789969 47076539613632 model_training_utils.py:505] Train Step: 1435/2100 / loss = 0.8623046875 I0421 10:59:04.790410 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 3534 and 3535 I0421 10:59:05.876456 47076539613632 model_training_utils.py:505] Train Step: 1436/2100 / loss = 0.5772705078125 I0421 10:59:05.876882 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.67 examples/second between steps 3535 and 3536 I0421 10:59:06.961900 47076539613632 model_training_utils.py:505] Train Step: 1437/2100 / loss = 0.905517578125 I0421 10:59:06.962330 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 3536 and 3537 I0421 10:59:08.046061 47076539613632 model_training_utils.py:505] Train Step: 1438/2100 / loss = 0.791259765625 I0421 10:59:08.046505 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 3537 and 3538 I0421 10:59:09.130537 47076539613632 model_training_utils.py:505] Train Step: 1439/2100 / loss = 1.24853515625 I0421 10:59:09.130974 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 3538 and 3539 I0421 10:59:10.211524 47076539613632 model_training_utils.py:505] Train Step: 1440/2100 / loss = 1.096923828125 I0421 10:59:10.211953 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 3539 and 3540 I0421 10:59:11.294247 47076539613632 model_training_utils.py:505] Train Step: 1441/2100 / loss = 1.01513671875 I0421 10:59:11.294676 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 3540 and 3541 I0421 10:59:12.380240 47076539613632 model_training_utils.py:505] Train Step: 1442/2100 / loss = 1.29736328125 I0421 10:59:12.380667 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 3541 and 3542 I0421 10:59:13.467035 47076539613632 model_training_utils.py:505] Train Step: 1443/2100 / loss = 1.271728515625 I0421 10:59:13.467467 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.64 examples/second between steps 3542 and 3543 I0421 10:59:14.557378 47076539613632 model_training_utils.py:505] Train Step: 1444/2100 / loss = 0.9873046875 I0421 10:59:14.557801 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.25 examples/second between steps 3543 and 3544 I0421 10:59:15.644727 47076539613632 model_training_utils.py:505] Train Step: 1445/2100 / loss = 0.77734375 I0421 10:59:15.645151 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.61 examples/second between steps 3544 and 3545 I0421 10:59:16.729630 47076539613632 model_training_utils.py:505] Train Step: 1446/2100 / loss = 0.9564208984375 I0421 10:59:16.730054 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 3545 and 3546 I0421 10:59:17.815092 47076539613632 model_training_utils.py:505] Train Step: 1447/2100 / loss = 0.843994140625 I0421 10:59:17.815530 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 3546 and 3547 I0421 10:59:18.897532 47076539613632 model_training_utils.py:505] Train Step: 1448/2100 / loss = 1.12841796875 I0421 10:59:18.897962 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 3547 and 3548 I0421 10:59:19.979218 47076539613632 model_training_utils.py:505] Train Step: 1449/2100 / loss = 0.9439697265625 I0421 10:59:19.979658 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 3548 and 3549 I0421 10:59:21.065090 47076539613632 model_training_utils.py:505] Train Step: 1450/2100 / loss = 1.0032958984375 I0421 10:59:21.065525 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 3549 and 3550 I0421 10:59:22.142330 47076539613632 model_training_utils.py:505] Train Step: 1451/2100 / loss = 1.34521484375 I0421 10:59:22.142762 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.70 examples/second between steps 3550 and 3551 I0421 10:59:23.227910 47076539613632 model_training_utils.py:505] Train Step: 1452/2100 / loss = 1.1661376953125 I0421 10:59:23.228343 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 3551 and 3552 I0421 10:59:24.312613 47076539613632 model_training_utils.py:505] Train Step: 1453/2100 / loss = 1.3720703125 I0421 10:59:24.313043 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 3552 and 3553 I0421 10:59:25.400325 47076539613632 model_training_utils.py:505] Train Step: 1454/2100 / loss = 1.262451171875 I0421 10:59:25.400755 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.55 examples/second between steps 3553 and 3554 I0421 10:59:26.492160 47076539613632 model_training_utils.py:505] Train Step: 1455/2100 / loss = 1.60302734375 I0421 10:59:26.492593 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.14 examples/second between steps 3554 and 3555 I0421 10:59:27.582455 47076539613632 model_training_utils.py:505] Train Step: 1456/2100 / loss = 1.153076171875 I0421 10:59:27.582874 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.32 examples/second between steps 3555 and 3556 I0421 10:59:28.669205 47076539613632 model_training_utils.py:505] Train Step: 1457/2100 / loss = 1.18017578125 I0421 10:59:28.669645 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 3556 and 3557 I0421 10:59:29.758631 47076539613632 model_training_utils.py:505] Train Step: 1458/2100 / loss = 1.1265869140625 I0421 10:59:29.759063 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.37 examples/second between steps 3557 and 3558 I0421 10:59:30.854995 47076539613632 model_training_utils.py:505] Train Step: 1459/2100 / loss = 0.83074951171875 I0421 10:59:30.855441 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.60 examples/second between steps 3558 and 3559 I0421 10:59:31.949709 47076539613632 model_training_utils.py:505] Train Step: 1460/2100 / loss = 0.71923828125 I0421 10:59:31.950126 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.77 examples/second between steps 3559 and 3560 I0421 10:59:33.040345 47076539613632 model_training_utils.py:505] Train Step: 1461/2100 / loss = 0.85302734375 I0421 10:59:33.040770 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.21 examples/second between steps 3560 and 3561 I0421 10:59:34.128337 47076539613632 model_training_utils.py:505] Train Step: 1462/2100 / loss = 1.1434326171875 I0421 10:59:34.128779 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.51 examples/second between steps 3561 and 3562 I0421 10:59:35.214703 47076539613632 model_training_utils.py:505] Train Step: 1463/2100 / loss = 1.22900390625 I0421 10:59:35.215129 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 3562 and 3563 I0421 10:59:36.303164 47076539613632 model_training_utils.py:505] Train Step: 1464/2100 / loss = 1.198974609375 I0421 10:59:36.303591 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.44 examples/second between steps 3563 and 3564 I0421 10:59:37.394016 47076539613632 model_training_utils.py:505] Train Step: 1465/2100 / loss = 1.175048828125 I0421 10:59:37.394446 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.19 examples/second between steps 3564 and 3565 I0421 10:59:38.480976 47076539613632 model_training_utils.py:505] Train Step: 1466/2100 / loss = 1.0689697265625 I0421 10:59:38.481404 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.63 examples/second between steps 3565 and 3566 I0421 10:59:39.563042 47076539613632 model_training_utils.py:505] Train Step: 1467/2100 / loss = 1.0921630859375 I0421 10:59:39.563477 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 3566 and 3567 I0421 10:59:40.649000 47076539613632 model_training_utils.py:505] Train Step: 1468/2100 / loss = 1.0028076171875 I0421 10:59:40.649428 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 3567 and 3568 I0421 10:59:41.735220 47076539613632 model_training_utils.py:505] Train Step: 1469/2100 / loss = 1.077392578125 I0421 10:59:41.735647 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.69 examples/second between steps 3568 and 3569 I0421 10:59:42.821303 47076539613632 model_training_utils.py:505] Train Step: 1470/2100 / loss = 1.2763671875 I0421 10:59:42.821728 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 3569 and 3570 I0421 10:59:43.908256 47076539613632 model_training_utils.py:505] Train Step: 1471/2100 / loss = 1.622314453125 I0421 10:59:43.908690 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 3570 and 3571 I0421 10:59:44.997932 47076539613632 model_training_utils.py:505] Train Step: 1472/2100 / loss = 1.739013671875 I0421 10:59:44.998373 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.30 examples/second between steps 3571 and 3572 I0421 10:59:46.086860 47076539613632 model_training_utils.py:505] Train Step: 1473/2100 / loss = 1.148193359375 I0421 10:59:46.087291 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.43 examples/second between steps 3572 and 3573 I0421 10:59:47.172894 47076539613632 model_training_utils.py:505] Train Step: 1474/2100 / loss = 1.0321044921875 I0421 10:59:47.173328 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 3573 and 3574 I0421 10:59:48.259286 47076539613632 model_training_utils.py:505] Train Step: 1475/2100 / loss = 0.83984375 I0421 10:59:48.259711 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 3574 and 3575 I0421 10:59:49.348323 47076539613632 model_training_utils.py:505] Train Step: 1476/2100 / loss = 0.8394775390625 I0421 10:59:49.348773 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.39 examples/second between steps 3575 and 3576 I0421 10:59:50.440548 47076539613632 model_training_utils.py:505] Train Step: 1477/2100 / loss = 0.82470703125 I0421 10:59:50.440975 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.07 examples/second between steps 3576 and 3577 I0421 10:59:51.533088 47076539613632 model_training_utils.py:505] Train Step: 1478/2100 / loss = 0.8912353515625 I0421 10:59:51.533519 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.05 examples/second between steps 3577 and 3578 I0421 10:59:52.633764 47076539613632 model_training_utils.py:505] Train Step: 1479/2100 / loss = 0.8011474609375 I0421 10:59:52.634191 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.13 examples/second between steps 3578 and 3579 I0421 10:59:53.729244 47076539613632 model_training_utils.py:505] Train Step: 1480/2100 / loss = 0.88916015625 I0421 10:59:53.729675 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.71 examples/second between steps 3579 and 3580 I0421 10:59:54.829416 47076539613632 model_training_utils.py:505] Train Step: 1481/2100 / loss = 0.9178466796875 I0421 10:59:54.829840 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.21 examples/second between steps 3580 and 3581 I0421 10:59:55.920290 47076539613632 model_training_utils.py:505] Train Step: 1482/2100 / loss = 1.10888671875 I0421 10:59:55.920717 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.20 examples/second between steps 3581 and 3582 I0421 10:59:57.015654 47076539613632 model_training_utils.py:505] Train Step: 1483/2100 / loss = 1.179443359375 I0421 10:59:57.016076 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.70 examples/second between steps 3582 and 3583 I0421 10:59:58.107416 47076539613632 model_training_utils.py:505] Train Step: 1484/2100 / loss = 1.40478515625 I0421 10:59:58.107836 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.12 examples/second between steps 3583 and 3584 I0421 10:59:59.194120 47076539613632 model_training_utils.py:505] Train Step: 1485/2100 / loss = 0.957275390625 I0421 10:59:59.194555 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.67 examples/second between steps 3584 and 3585 I0421 11:00:00.286663 47076539613632 model_training_utils.py:505] Train Step: 1486/2100 / loss = 1.37890625 I0421 11:00:00.287089 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.02 examples/second between steps 3585 and 3586 I0421 11:00:01.376600 47076539613632 model_training_utils.py:505] Train Step: 1487/2100 / loss = 1.268798828125 I0421 11:00:01.377031 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.31 examples/second between steps 3586 and 3587 I0421 11:00:02.470187 47076539613632 model_training_utils.py:505] Train Step: 1488/2100 / loss = 1.1231689453125 I0421 11:00:02.470614 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.92 examples/second between steps 3587 and 3588 I0421 11:00:03.565191 47076539613632 model_training_utils.py:505] Train Step: 1489/2100 / loss = 1.21337890625 I0421 11:00:03.565626 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.73 examples/second between steps 3588 and 3589 I0421 11:00:04.657159 47076539613632 model_training_utils.py:505] Train Step: 1490/2100 / loss = 0.99853515625 I0421 11:00:04.657591 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.07 examples/second between steps 3589 and 3590 I0421 11:00:05.738060 47076539613632 model_training_utils.py:505] Train Step: 1491/2100 / loss = 0.8560791015625 I0421 11:00:05.738494 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.29 examples/second between steps 3590 and 3591 I0421 11:00:06.821704 47076539613632 model_training_utils.py:505] Train Step: 1492/2100 / loss = 0.9356689453125 I0421 11:00:06.822138 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 3591 and 3592 I0421 11:00:07.905073 47076539613632 model_training_utils.py:505] Train Step: 1493/2100 / loss = 0.5985107421875 I0421 11:00:07.905512 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 3592 and 3593 I0421 11:00:08.987782 47076539613632 model_training_utils.py:505] Train Step: 1494/2100 / loss = 0.934326171875 I0421 11:00:08.988214 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 3593 and 3594 I0421 11:00:10.074200 47076539613632 model_training_utils.py:505] Train Step: 1495/2100 / loss = 0.760498046875 I0421 11:00:10.074634 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 3594 and 3595 I0421 11:00:11.158252 47076539613632 model_training_utils.py:505] Train Step: 1496/2100 / loss = 0.9112548828125 I0421 11:00:11.158687 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 3595 and 3596 I0421 11:00:12.237002 47076539613632 model_training_utils.py:505] Train Step: 1497/2100 / loss = 0.8759765625 I0421 11:00:12.237430 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.51 examples/second between steps 3596 and 3597 I0421 11:00:13.319244 47076539613632 model_training_utils.py:505] Train Step: 1498/2100 / loss = 0.8927001953125 I0421 11:00:13.319678 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 3597 and 3598 I0421 11:00:14.401967 47076539613632 model_training_utils.py:505] Train Step: 1499/2100 / loss = 0.9072265625 I0421 11:00:14.402398 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 3598 and 3599 I0421 11:00:15.488664 47076539613632 model_training_utils.py:505] Train Step: 1500/2100 / loss = 1.0191650390625 I0421 11:00:15.489097 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 3599 and 3600 I0421 11:00:16.578193 47076539613632 model_training_utils.py:505] Train Step: 1501/2100 / loss = 0.97412109375 I0421 11:00:16.578629 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.33 examples/second between steps 3600 and 3601 I0421 11:00:17.658270 47076539613632 model_training_utils.py:505] Train Step: 1502/2100 / loss = 0.79443359375 I0421 11:00:17.658707 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.40 examples/second between steps 3601 and 3602 I0421 11:00:18.745313 47076539613632 model_training_utils.py:505] Train Step: 1503/2100 / loss = 0.7080078125 I0421 11:00:18.745738 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.61 examples/second between steps 3602 and 3603 I0421 11:00:19.836706 47076539613632 model_training_utils.py:505] Train Step: 1504/2100 / loss = 0.7574462890625 I0421 11:00:19.837140 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.13 examples/second between steps 3603 and 3604 I0421 11:00:20.928722 47076539613632 model_training_utils.py:505] Train Step: 1505/2100 / loss = 1.249267578125 I0421 11:00:20.929161 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.08 examples/second between steps 3604 and 3605 I0421 11:00:22.021761 47076539613632 model_training_utils.py:505] Train Step: 1506/2100 / loss = 0.981201171875 I0421 11:00:22.022187 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 117.98 examples/second between steps 3605 and 3606 I0421 11:00:23.115723 47076539613632 model_training_utils.py:505] Train Step: 1507/2100 / loss = 0.784423828125 I0421 11:00:23.116146 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.86 examples/second between steps 3606 and 3607 I0421 11:00:24.204802 47076539613632 model_training_utils.py:505] Train Step: 1508/2100 / loss = 0.7537841796875 I0421 11:00:24.205229 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.41 examples/second between steps 3607 and 3608 I0421 11:00:25.293177 47076539613632 model_training_utils.py:505] Train Step: 1509/2100 / loss = 0.7462158203125 I0421 11:00:25.293616 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.49 examples/second between steps 3608 and 3609 I0421 11:00:26.381365 47076539613632 model_training_utils.py:505] Train Step: 1510/2100 / loss = 0.9361572265625 I0421 11:00:26.381802 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.50 examples/second between steps 3609 and 3610 I0421 11:00:27.470993 47076539613632 model_training_utils.py:505] Train Step: 1511/2100 / loss = 0.9716796875 I0421 11:00:27.471433 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.35 examples/second between steps 3610 and 3611 I0421 11:00:28.563398 47076539613632 model_training_utils.py:505] Train Step: 1512/2100 / loss = 1.035888671875 I0421 11:00:28.563822 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.12 examples/second between steps 3611 and 3612 I0421 11:00:29.655464 47076539613632 model_training_utils.py:505] Train Step: 1513/2100 / loss = 0.97509765625 I0421 11:00:29.655894 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.07 examples/second between steps 3612 and 3613 I0421 11:00:30.747677 47076539613632 model_training_utils.py:505] Train Step: 1514/2100 / loss = 0.858642578125 I0421 11:00:30.748106 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.06 examples/second between steps 3613 and 3614 I0421 11:00:31.839817 47076539613632 model_training_utils.py:505] Train Step: 1515/2100 / loss = 0.63165283203125 I0421 11:00:31.840245 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.06 examples/second between steps 3614 and 3615 I0421 11:00:32.940382 47076539613632 model_training_utils.py:505] Train Step: 1516/2100 / loss = 0.9415283203125 I0421 11:00:32.940798 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.16 examples/second between steps 3615 and 3616 I0421 11:00:34.034798 47076539613632 model_training_utils.py:505] Train Step: 1517/2100 / loss = 1.29443359375 I0421 11:00:34.035224 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.81 examples/second between steps 3616 and 3617 I0421 11:00:35.131111 47076539613632 model_training_utils.py:505] Train Step: 1518/2100 / loss = 0.955810546875 I0421 11:00:35.131545 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.61 examples/second between steps 3617 and 3618 I0421 11:00:36.215837 47076539613632 model_training_utils.py:505] Train Step: 1519/2100 / loss = 0.803466796875 I0421 11:00:36.216258 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 3618 and 3619 I0421 11:00:37.301711 47076539613632 model_training_utils.py:505] Train Step: 1520/2100 / loss = 0.9547119140625 I0421 11:00:37.302133 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 3619 and 3620 I0421 11:00:38.388353 47076539613632 model_training_utils.py:505] Train Step: 1521/2100 / loss = 0.940185546875 I0421 11:00:38.388781 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.67 examples/second between steps 3620 and 3621 I0421 11:00:39.480132 47076539613632 model_training_utils.py:505] Train Step: 1522/2100 / loss = 1.0054931640625 I0421 11:00:39.480580 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.11 examples/second between steps 3621 and 3622 I0421 11:00:40.571974 47076539613632 model_training_utils.py:505] Train Step: 1523/2100 / loss = 1.14404296875 I0421 11:00:40.572405 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.11 examples/second between steps 3622 and 3623 I0421 11:00:41.667691 47076539613632 model_training_utils.py:505] Train Step: 1524/2100 / loss = 1.09814453125 I0421 11:00:41.668114 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.69 examples/second between steps 3623 and 3624 I0421 11:00:42.762988 47076539613632 model_training_utils.py:505] Train Step: 1525/2100 / loss = 0.9949951171875 I0421 11:00:42.763428 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.74 examples/second between steps 3624 and 3625 I0421 11:00:43.863153 47076539613632 model_training_utils.py:505] Train Step: 1526/2100 / loss = 1.124267578125 I0421 11:00:43.863601 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.21 examples/second between steps 3625 and 3626 I0421 11:00:44.960239 47076539613632 model_training_utils.py:505] Train Step: 1527/2100 / loss = 0.9730224609375 I0421 11:00:44.960681 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.51 examples/second between steps 3626 and 3627 I0421 11:00:46.056225 47076539613632 model_training_utils.py:505] Train Step: 1528/2100 / loss = 1.25537109375 I0421 11:00:46.056658 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.63 examples/second between steps 3627 and 3628 I0421 11:00:47.146404 47076539613632 model_training_utils.py:505] Train Step: 1529/2100 / loss = 1.2535400390625 I0421 11:00:47.146830 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.26 examples/second between steps 3628 and 3629 I0421 11:00:48.232882 47076539613632 model_training_utils.py:505] Train Step: 1530/2100 / loss = 0.8759765625 I0421 11:00:48.233315 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 3629 and 3630 I0421 11:00:49.323352 47076539613632 model_training_utils.py:505] Train Step: 1531/2100 / loss = 0.821044921875 I0421 11:00:49.323782 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.23 examples/second between steps 3630 and 3631 I0421 11:00:50.414425 47076539613632 model_training_utils.py:505] Train Step: 1532/2100 / loss = 1.0296630859375 I0421 11:00:50.414849 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.18 examples/second between steps 3631 and 3632 I0421 11:00:51.503708 47076539613632 model_training_utils.py:505] Train Step: 1533/2100 / loss = 0.946533203125 I0421 11:00:51.504133 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.36 examples/second between steps 3632 and 3633 I0421 11:00:52.592601 47076539613632 model_training_utils.py:505] Train Step: 1534/2100 / loss = 0.80517578125 I0421 11:00:52.593023 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.40 examples/second between steps 3633 and 3634 I0421 11:00:53.677755 47076539613632 model_training_utils.py:505] Train Step: 1535/2100 / loss = 1.015380859375 I0421 11:00:53.678181 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 3634 and 3635 I0421 11:00:54.761707 47076539613632 model_training_utils.py:505] Train Step: 1536/2100 / loss = 1.0264892578125 I0421 11:00:54.762133 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 3635 and 3636 I0421 11:00:55.852350 47076539613632 model_training_utils.py:505] Train Step: 1537/2100 / loss = 0.956298828125 I0421 11:00:55.852773 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.24 examples/second between steps 3636 and 3637 I0421 11:00:56.945510 47076539613632 model_training_utils.py:505] Train Step: 1538/2100 / loss = 0.960693359375 I0421 11:00:56.945934 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.94 examples/second between steps 3637 and 3638 I0421 11:00:58.034519 47076539613632 model_training_utils.py:505] Train Step: 1539/2100 / loss = 1.0169677734375 I0421 11:00:58.034957 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.43 examples/second between steps 3638 and 3639 I0421 11:00:59.126833 47076539613632 model_training_utils.py:505] Train Step: 1540/2100 / loss = 0.8065185546875 I0421 11:00:59.127264 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.02 examples/second between steps 3639 and 3640 I0421 11:01:00.222470 47076539613632 model_training_utils.py:505] Train Step: 1541/2100 / loss = 1.04443359375 I0421 11:01:00.222899 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.67 examples/second between steps 3640 and 3641 I0421 11:01:01.314956 47076539613632 model_training_utils.py:505] Train Step: 1542/2100 / loss = 0.9561767578125 I0421 11:01:01.315408 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 117.99 examples/second between steps 3641 and 3642 I0421 11:01:02.407991 47076539613632 model_training_utils.py:505] Train Step: 1543/2100 / loss = 0.998779296875 I0421 11:01:02.408428 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.94 examples/second between steps 3642 and 3643 I0421 11:01:03.504026 47076539613632 model_training_utils.py:505] Train Step: 1544/2100 / loss = 0.95703125 I0421 11:01:03.504459 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.60 examples/second between steps 3643 and 3644 I0421 11:01:04.596188 47076539613632 model_training_utils.py:505] Train Step: 1545/2100 / loss = 0.7236328125 I0421 11:01:04.596618 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.04 examples/second between steps 3644 and 3645 I0421 11:01:05.681549 47076539613632 model_training_utils.py:505] Train Step: 1546/2100 / loss = 1.3043212890625 I0421 11:01:05.681978 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 3645 and 3646 I0421 11:01:06.769630 47076539613632 model_training_utils.py:505] Train Step: 1547/2100 / loss = 0.82568359375 I0421 11:01:06.770061 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.49 examples/second between steps 3646 and 3647 I0421 11:01:07.851777 47076539613632 model_training_utils.py:505] Train Step: 1548/2100 / loss = 0.78125 I0421 11:01:07.852204 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 3647 and 3648 I0421 11:01:08.939822 47076539613632 model_training_utils.py:505] Train Step: 1549/2100 / loss = 0.69970703125 I0421 11:01:08.940248 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.51 examples/second between steps 3648 and 3649 I0421 11:01:10.022566 47076539613632 model_training_utils.py:505] Train Step: 1550/2100 / loss = 0.8369140625 I0421 11:01:10.022994 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.07 examples/second between steps 3649 and 3650 I0421 11:01:11.103845 47076539613632 model_training_utils.py:505] Train Step: 1551/2100 / loss = 0.58447265625 I0421 11:01:11.104274 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 3650 and 3651 I0421 11:01:12.191371 47076539613632 model_training_utils.py:505] Train Step: 1552/2100 / loss = 0.4427490234375 I0421 11:01:12.191801 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.56 examples/second between steps 3651 and 3652 I0421 11:01:13.278551 47076539613632 model_training_utils.py:505] Train Step: 1553/2100 / loss = 0.51904296875 I0421 11:01:13.278980 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.61 examples/second between steps 3652 and 3653 I0421 11:01:14.368803 47076539613632 model_training_utils.py:505] Train Step: 1554/2100 / loss = 0.550048828125 I0421 11:01:14.369230 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.26 examples/second between steps 3653 and 3654 I0421 11:01:15.457479 47076539613632 model_training_utils.py:505] Train Step: 1555/2100 / loss = 0.880859375 I0421 11:01:15.457903 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.45 examples/second between steps 3654 and 3655 I0421 11:01:16.547391 47076539613632 model_training_utils.py:505] Train Step: 1556/2100 / loss = 0.9951171875 I0421 11:01:16.547817 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.32 examples/second between steps 3655 and 3656 I0421 11:01:17.628580 47076539613632 model_training_utils.py:505] Train Step: 1557/2100 / loss = 0.898681640625 I0421 11:01:17.629006 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.27 examples/second between steps 3656 and 3657 I0421 11:01:18.711537 47076539613632 model_training_utils.py:505] Train Step: 1558/2100 / loss = 1.0458984375 I0421 11:01:18.711962 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 3657 and 3658 I0421 11:01:19.793144 47076539613632 model_training_utils.py:505] Train Step: 1559/2100 / loss = 0.7569580078125 I0421 11:01:19.793579 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 3658 and 3659 I0421 11:01:20.884674 47076539613632 model_training_utils.py:505] Train Step: 1560/2100 / loss = 0.7607421875 I0421 11:01:20.885094 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.12 examples/second between steps 3659 and 3660 I0421 11:01:21.976319 47076539613632 model_training_utils.py:505] Train Step: 1561/2100 / loss = 0.764892578125 I0421 11:01:21.976743 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.15 examples/second between steps 3660 and 3661 I0421 11:01:23.066082 47076539613632 model_training_utils.py:505] Train Step: 1562/2100 / loss = 1.2606201171875 I0421 11:01:23.066508 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.29 examples/second between steps 3661 and 3662 I0421 11:01:24.157801 47076539613632 model_training_utils.py:505] Train Step: 1563/2100 / loss = 1.390380859375 I0421 11:01:24.158232 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.10 examples/second between steps 3662 and 3663 I0421 11:01:25.237934 47076539613632 model_training_utils.py:505] Train Step: 1564/2100 / loss = 0.998046875 I0421 11:01:25.238364 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.38 examples/second between steps 3663 and 3664 I0421 11:01:26.319208 47076539613632 model_training_utils.py:505] Train Step: 1565/2100 / loss = 0.8743896484375 I0421 11:01:26.319636 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.29 examples/second between steps 3664 and 3665 I0421 11:01:27.404943 47076539613632 model_training_utils.py:505] Train Step: 1566/2100 / loss = 0.7315673828125 I0421 11:01:27.405375 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 3665 and 3666 I0421 11:01:28.491720 47076539613632 model_training_utils.py:505] Train Step: 1567/2100 / loss = 0.5479736328125 I0421 11:01:28.492140 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 3666 and 3667 I0421 11:01:29.581218 47076539613632 model_training_utils.py:505] Train Step: 1568/2100 / loss = 1.1820068359375 I0421 11:01:29.581649 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.34 examples/second between steps 3667 and 3668 I0421 11:01:30.671420 47076539613632 model_training_utils.py:505] Train Step: 1569/2100 / loss = 0.92822265625 I0421 11:01:30.671844 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.26 examples/second between steps 3668 and 3669 I0421 11:01:31.764385 47076539613632 model_training_utils.py:505] Train Step: 1570/2100 / loss = 0.5933837890625 I0421 11:01:31.764814 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.97 examples/second between steps 3669 and 3670 I0421 11:01:32.859856 47076539613632 model_training_utils.py:505] Train Step: 1571/2100 / loss = 0.6630859375 I0421 11:01:32.860290 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.70 examples/second between steps 3670 and 3671 I0421 11:01:33.952518 47076539613632 model_training_utils.py:505] Train Step: 1572/2100 / loss = 0.6558837890625 I0421 11:01:33.952943 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.00 examples/second between steps 3671 and 3672 I0421 11:01:35.045765 47076539613632 model_training_utils.py:505] Train Step: 1573/2100 / loss = 0.7950439453125 I0421 11:01:35.046200 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.95 examples/second between steps 3672 and 3673 I0421 11:01:36.126659 47076539613632 model_training_utils.py:505] Train Step: 1574/2100 / loss = 0.968505859375 I0421 11:01:36.127090 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.30 examples/second between steps 3673 and 3674 I0421 11:01:37.208400 47076539613632 model_training_utils.py:505] Train Step: 1575/2100 / loss = 0.71124267578125 I0421 11:01:37.208832 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 3674 and 3675 I0421 11:01:38.295868 47076539613632 model_training_utils.py:505] Train Step: 1576/2100 / loss = 0.974853515625 I0421 11:01:38.296307 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.57 examples/second between steps 3675 and 3676 I0421 11:01:39.380324 47076539613632 model_training_utils.py:505] Train Step: 1577/2100 / loss = 0.68218994140625 I0421 11:01:39.380753 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 3676 and 3677 I0421 11:01:40.459958 47076539613632 model_training_utils.py:505] Train Step: 1578/2100 / loss = 0.5631103515625 I0421 11:01:40.460391 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.42 examples/second between steps 3677 and 3678 I0421 11:01:41.552976 47076539613632 model_training_utils.py:505] Train Step: 1579/2100 / loss = 0.838623046875 I0421 11:01:41.553422 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.94 examples/second between steps 3678 and 3679 I0421 11:01:42.640502 47076539613632 model_training_utils.py:505] Train Step: 1580/2100 / loss = 0.74786376953125 I0421 11:01:42.640919 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.55 examples/second between steps 3679 and 3680 I0421 11:01:43.735543 47076539613632 model_training_utils.py:505] Train Step: 1581/2100 / loss = 0.9508056640625 I0421 11:01:43.735963 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.70 examples/second between steps 3680 and 3681 I0421 11:01:44.832424 47076539613632 model_training_utils.py:505] Train Step: 1582/2100 / loss = 1.04833984375 I0421 11:01:44.832844 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.59 examples/second between steps 3681 and 3682 I0421 11:01:45.925513 47076539613632 model_training_utils.py:505] Train Step: 1583/2100 / loss = 0.8880615234375 I0421 11:01:45.925935 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.92 examples/second between steps 3682 and 3683 I0421 11:01:47.014402 47076539613632 model_training_utils.py:505] Train Step: 1584/2100 / loss = 0.79248046875 I0421 11:01:47.014836 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.43 examples/second between steps 3683 and 3684 I0421 11:01:48.102769 47076539613632 model_training_utils.py:505] Train Step: 1585/2100 / loss = 0.7679443359375 I0421 11:01:48.103193 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.48 examples/second between steps 3684 and 3685 I0421 11:01:49.190809 47076539613632 model_training_utils.py:505] Train Step: 1586/2100 / loss = 0.705078125 I0421 11:01:49.191242 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.52 examples/second between steps 3685 and 3686 I0421 11:01:50.274002 47076539613632 model_training_utils.py:505] Train Step: 1587/2100 / loss = 0.6142578125 I0421 11:01:50.274440 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 3686 and 3687 I0421 11:01:51.360168 47076539613632 model_training_utils.py:505] Train Step: 1588/2100 / loss = 0.7249755859375 I0421 11:01:51.360605 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 3687 and 3688 I0421 11:01:52.445415 47076539613632 model_training_utils.py:505] Train Step: 1589/2100 / loss = 0.6414794921875 I0421 11:01:52.445843 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 3688 and 3689 I0421 11:01:53.535185 47076539613632 model_training_utils.py:505] Train Step: 1590/2100 / loss = 0.6212158203125 I0421 11:01:53.535614 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.40 examples/second between steps 3689 and 3690 I0421 11:01:54.629105 47076539613632 model_training_utils.py:505] Train Step: 1591/2100 / loss = 0.8892822265625 I0421 11:01:54.629537 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.89 examples/second between steps 3690 and 3691 I0421 11:01:55.724843 47076539613632 model_training_utils.py:505] Train Step: 1592/2100 / loss = 1.366943359375 I0421 11:01:55.725260 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.70 examples/second between steps 3691 and 3692 I0421 11:01:56.813372 47076539613632 model_training_utils.py:505] Train Step: 1593/2100 / loss = 1.3798828125 I0421 11:01:56.813780 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.48 examples/second between steps 3692 and 3693 I0421 11:01:57.899310 47076539613632 model_training_utils.py:505] Train Step: 1594/2100 / loss = 0.9927978515625 I0421 11:01:57.899728 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 3693 and 3694 I0421 11:01:58.991501 47076539613632 model_training_utils.py:505] Train Step: 1595/2100 / loss = 0.951171875 I0421 11:01:58.991929 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.08 examples/second between steps 3694 and 3695 I0421 11:02:00.078921 47076539613632 model_training_utils.py:505] Train Step: 1596/2100 / loss = 0.7161865234375 I0421 11:02:00.079353 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 3695 and 3696 I0421 11:02:01.170853 47076539613632 model_training_utils.py:505] Train Step: 1597/2100 / loss = 0.6243896484375 I0421 11:02:01.171294 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.11 examples/second between steps 3696 and 3697 I0421 11:02:02.252511 47076539613632 model_training_utils.py:505] Train Step: 1598/2100 / loss = 0.6832275390625 I0421 11:02:02.252937 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 3697 and 3698 I0421 11:02:03.347535 47076539613632 model_training_utils.py:505] Train Step: 1599/2100 / loss = 1.047119140625 I0421 11:02:03.347962 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.82 examples/second between steps 3698 and 3699 I0421 11:02:04.439790 47076539613632 model_training_utils.py:505] Train Step: 1600/2100 / loss = 0.7828369140625 I0421 11:02:04.440222 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.06 examples/second between steps 3699 and 3700 I0421 11:02:05.522553 47076539613632 model_training_utils.py:505] Train Step: 1601/2100 / loss = 0.64453125 I0421 11:02:05.522986 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 3700 and 3701 I0421 11:02:06.609549 47076539613632 model_training_utils.py:505] Train Step: 1602/2100 / loss = 0.7745361328125 I0421 11:02:06.609956 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.68 examples/second between steps 3701 and 3702 I0421 11:02:07.697337 47076539613632 model_training_utils.py:505] Train Step: 1603/2100 / loss = 0.987548828125 I0421 11:02:07.697754 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.55 examples/second between steps 3702 and 3703 I0421 11:02:08.787536 47076539613632 model_training_utils.py:505] Train Step: 1604/2100 / loss = 0.82421875 I0421 11:02:08.787960 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.30 examples/second between steps 3703 and 3704 I0421 11:02:09.877907 47076539613632 model_training_utils.py:505] Train Step: 1605/2100 / loss = 0.88818359375 I0421 11:02:09.878339 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.32 examples/second between steps 3704 and 3705 I0421 11:02:10.963121 47076539613632 model_training_utils.py:505] Train Step: 1606/2100 / loss = 0.6893310546875 I0421 11:02:10.963550 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 3705 and 3706 I0421 11:02:12.047021 47076539613632 model_training_utils.py:505] Train Step: 1607/2100 / loss = 0.9373779296875 I0421 11:02:12.047469 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 3706 and 3707 I0421 11:02:13.128166 47076539613632 model_training_utils.py:505] Train Step: 1608/2100 / loss = 0.630615234375 I0421 11:02:13.128615 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.29 examples/second between steps 3707 and 3708 I0421 11:02:14.211710 47076539613632 model_training_utils.py:505] Train Step: 1609/2100 / loss = 0.683349609375 I0421 11:02:14.212133 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 3708 and 3709 I0421 11:02:15.292767 47076539613632 model_training_utils.py:505] Train Step: 1610/2100 / loss = 0.5587158203125 I0421 11:02:15.293190 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 3709 and 3710 I0421 11:02:16.375311 47076539613632 model_training_utils.py:505] Train Step: 1611/2100 / loss = 0.7322998046875 I0421 11:02:16.375741 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 3710 and 3711 I0421 11:02:17.458427 47076539613632 model_training_utils.py:505] Train Step: 1612/2100 / loss = 0.712890625 I0421 11:02:17.458869 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 3711 and 3712 I0421 11:02:18.540647 47076539613632 model_training_utils.py:505] Train Step: 1613/2100 / loss = 0.9847412109375 I0421 11:02:18.541074 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 3712 and 3713 I0421 11:02:19.625357 47076539613632 model_training_utils.py:505] Train Step: 1614/2100 / loss = 0.74267578125 I0421 11:02:19.625788 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 3713 and 3714 I0421 11:02:20.710021 47076539613632 model_training_utils.py:505] Train Step: 1615/2100 / loss = 0.972900390625 I0421 11:02:20.710455 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 3714 and 3715 I0421 11:02:21.794237 47076539613632 model_training_utils.py:505] Train Step: 1616/2100 / loss = 0.9776611328125 I0421 11:02:21.794675 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 3715 and 3716 I0421 11:02:22.875413 47076539613632 model_training_utils.py:505] Train Step: 1617/2100 / loss = 0.7540283203125 I0421 11:02:22.875844 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.30 examples/second between steps 3716 and 3717 I0421 11:02:23.962473 47076539613632 model_training_utils.py:505] Train Step: 1618/2100 / loss = 0.9359130859375 I0421 11:02:23.962902 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.65 examples/second between steps 3717 and 3718 I0421 11:02:25.046388 47076539613632 model_training_utils.py:505] Train Step: 1619/2100 / loss = 0.839111328125 I0421 11:02:25.046833 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 3718 and 3719 I0421 11:02:26.131870 47076539613632 model_training_utils.py:505] Train Step: 1620/2100 / loss = 0.8524169921875 I0421 11:02:26.132311 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 3719 and 3720 I0421 11:02:27.214536 47076539613632 model_training_utils.py:505] Train Step: 1621/2100 / loss = 0.6749267578125 I0421 11:02:27.214977 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 3720 and 3721 I0421 11:02:28.296928 47076539613632 model_training_utils.py:505] Train Step: 1622/2100 / loss = 0.65765380859375 I0421 11:02:28.297370 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 3721 and 3722 I0421 11:02:29.378582 47076539613632 model_training_utils.py:505] Train Step: 1623/2100 / loss = 0.9283447265625 I0421 11:02:29.379017 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 3722 and 3723 I0421 11:02:30.461826 47076539613632 model_training_utils.py:505] Train Step: 1624/2100 / loss = 0.922607421875 I0421 11:02:30.462254 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 3723 and 3724 I0421 11:02:31.547856 47076539613632 model_training_utils.py:505] Train Step: 1625/2100 / loss = 0.640380859375 I0421 11:02:31.548297 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.75 examples/second between steps 3724 and 3725 I0421 11:02:32.635331 47076539613632 model_training_utils.py:505] Train Step: 1626/2100 / loss = 0.5872802734375 I0421 11:02:32.635760 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.56 examples/second between steps 3725 and 3726 I0421 11:02:33.720948 47076539613632 model_training_utils.py:505] Train Step: 1627/2100 / loss = 0.539306640625 I0421 11:02:33.721387 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 3726 and 3727 I0421 11:02:34.803167 47076539613632 model_training_utils.py:505] Train Step: 1628/2100 / loss = 0.73095703125 I0421 11:02:34.803602 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 3727 and 3728 I0421 11:02:35.888332 47076539613632 model_training_utils.py:505] Train Step: 1629/2100 / loss = 0.8878173828125 I0421 11:02:35.888752 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 3728 and 3729 I0421 11:02:36.971191 47076539613632 model_training_utils.py:505] Train Step: 1630/2100 / loss = 0.8197021484375 I0421 11:02:36.971623 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 3729 and 3730 I0421 11:02:38.053631 47076539613632 model_training_utils.py:505] Train Step: 1631/2100 / loss = 0.81201171875 I0421 11:02:38.054048 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 3730 and 3731 I0421 11:02:39.138470 47076539613632 model_training_utils.py:505] Train Step: 1632/2100 / loss = 0.6568603515625 I0421 11:02:39.138902 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 3731 and 3732 I0421 11:02:40.218867 47076539613632 model_training_utils.py:505] Train Step: 1633/2100 / loss = 0.8328857421875 I0421 11:02:40.219300 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.38 examples/second between steps 3732 and 3733 I0421 11:02:41.304627 47076539613632 model_training_utils.py:505] Train Step: 1634/2100 / loss = 1.127685546875 I0421 11:02:41.305057 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 3733 and 3734 I0421 11:02:42.386735 47076539613632 model_training_utils.py:505] Train Step: 1635/2100 / loss = 1.2457275390625 I0421 11:02:42.387171 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 3734 and 3735 I0421 11:02:43.467748 47076539613632 model_training_utils.py:505] Train Step: 1636/2100 / loss = 1.106201171875 I0421 11:02:43.468174 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.35 examples/second between steps 3735 and 3736 I0421 11:02:44.548243 47076539613632 model_training_utils.py:505] Train Step: 1637/2100 / loss = 1.0115966796875 I0421 11:02:44.548687 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.36 examples/second between steps 3736 and 3737 I0421 11:02:45.625048 47076539613632 model_training_utils.py:505] Train Step: 1638/2100 / loss = 0.7960205078125 I0421 11:02:45.625482 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.82 examples/second between steps 3737 and 3738 I0421 11:02:46.707399 47076539613632 model_training_utils.py:505] Train Step: 1639/2100 / loss = 0.713134765625 I0421 11:02:46.707828 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 3738 and 3739 I0421 11:02:47.790359 47076539613632 model_training_utils.py:505] Train Step: 1640/2100 / loss = 0.85205078125 I0421 11:02:47.790782 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 3739 and 3740 I0421 11:02:48.872711 47076539613632 model_training_utils.py:505] Train Step: 1641/2100 / loss = 1.0765380859375 I0421 11:02:48.873148 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 3740 and 3741 I0421 11:02:49.956662 47076539613632 model_training_utils.py:505] Train Step: 1642/2100 / loss = 1.26708984375 I0421 11:02:49.957094 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 3741 and 3742 I0421 11:02:51.039253 47076539613632 model_training_utils.py:505] Train Step: 1643/2100 / loss = 1.236572265625 I0421 11:02:51.039697 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 3742 and 3743 I0421 11:02:52.122017 47076539613632 model_training_utils.py:505] Train Step: 1644/2100 / loss = 1.037353515625 I0421 11:02:52.122453 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 3743 and 3744 I0421 11:02:53.205387 47076539613632 model_training_utils.py:505] Train Step: 1645/2100 / loss = 1.096435546875 I0421 11:02:53.205809 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 3744 and 3745 I0421 11:02:54.287472 47076539613632 model_training_utils.py:505] Train Step: 1646/2100 / loss = 1.188720703125 I0421 11:02:54.287913 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 3745 and 3746 I0421 11:02:55.367241 47076539613632 model_training_utils.py:505] Train Step: 1647/2100 / loss = 1.139892578125 I0421 11:02:55.367677 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.42 examples/second between steps 3746 and 3747 I0421 11:02:56.450354 47076539613632 model_training_utils.py:505] Train Step: 1648/2100 / loss = 0.898193359375 I0421 11:02:56.450788 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.07 examples/second between steps 3747 and 3748 I0421 11:02:57.530998 47076539613632 model_training_utils.py:505] Train Step: 1649/2100 / loss = 1.0625 I0421 11:02:57.531441 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.35 examples/second between steps 3748 and 3749 I0421 11:02:58.610026 47076539613632 model_training_utils.py:505] Train Step: 1650/2100 / loss = 0.6214599609375 I0421 11:02:58.610462 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.52 examples/second between steps 3749 and 3750 I0421 11:02:59.694117 47076539613632 model_training_utils.py:505] Train Step: 1651/2100 / loss = 0.6802978515625 I0421 11:02:59.694558 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 3750 and 3751 I0421 11:03:00.775241 47076539613632 model_training_utils.py:505] Train Step: 1652/2100 / loss = 0.52178955078125 I0421 11:03:00.775676 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 3751 and 3752 I0421 11:03:01.858080 47076539613632 model_training_utils.py:505] Train Step: 1653/2100 / loss = 0.5614013671875 I0421 11:03:01.858519 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 3752 and 3753 I0421 11:03:02.935235 47076539613632 model_training_utils.py:505] Train Step: 1654/2100 / loss = 0.50726318359375 I0421 11:03:02.935672 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.74 examples/second between steps 3753 and 3754 I0421 11:03:04.016144 47076539613632 model_training_utils.py:505] Train Step: 1655/2100 / loss = 0.6258544921875 I0421 11:03:04.016579 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.31 examples/second between steps 3754 and 3755 I0421 11:03:05.102286 47076539613632 model_training_utils.py:505] Train Step: 1656/2100 / loss = 0.809814453125 I0421 11:03:05.102727 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 3755 and 3756 I0421 11:03:06.185937 47076539613632 model_training_utils.py:505] Train Step: 1657/2100 / loss = 0.682861328125 I0421 11:03:06.186367 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 3756 and 3757 I0421 11:03:07.271713 47076539613632 model_training_utils.py:505] Train Step: 1658/2100 / loss = 0.6778564453125 I0421 11:03:07.272143 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 3757 and 3758 I0421 11:03:08.359383 47076539613632 model_training_utils.py:505] Train Step: 1659/2100 / loss = 0.683837890625 I0421 11:03:08.359814 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.56 examples/second between steps 3758 and 3759 I0421 11:03:09.444048 47076539613632 model_training_utils.py:505] Train Step: 1660/2100 / loss = 0.7003173828125 I0421 11:03:09.444493 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 3759 and 3760 I0421 11:03:10.523301 47076539613632 model_training_utils.py:505] Train Step: 1661/2100 / loss = 0.8477783203125 I0421 11:03:10.523729 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.50 examples/second between steps 3760 and 3761 I0421 11:03:11.606425 47076539613632 model_training_utils.py:505] Train Step: 1662/2100 / loss = 0.510498046875 I0421 11:03:11.606856 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 3761 and 3762 I0421 11:03:12.691155 47076539613632 model_training_utils.py:505] Train Step: 1663/2100 / loss = 0.45977783203125 I0421 11:03:12.691608 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 3762 and 3763 I0421 11:03:13.773006 47076539613632 model_training_utils.py:505] Train Step: 1664/2100 / loss = 0.4486083984375 I0421 11:03:13.773440 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 3763 and 3764 I0421 11:03:14.853455 47076539613632 model_training_utils.py:505] Train Step: 1665/2100 / loss = 0.8408203125 I0421 11:03:14.853877 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.35 examples/second between steps 3764 and 3765 I0421 11:03:15.937742 47076539613632 model_training_utils.py:505] Train Step: 1666/2100 / loss = 0.75390625 I0421 11:03:15.938171 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 3765 and 3766 I0421 11:03:17.022009 47076539613632 model_training_utils.py:505] Train Step: 1667/2100 / loss = 0.7333984375 I0421 11:03:17.022452 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 3766 and 3767 I0421 11:03:18.104538 47076539613632 model_training_utils.py:505] Train Step: 1668/2100 / loss = 1.03125 I0421 11:03:18.104967 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 3767 and 3768 I0421 11:03:19.192579 47076539613632 model_training_utils.py:505] Train Step: 1669/2100 / loss = 0.6396484375 I0421 11:03:19.193012 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.56 examples/second between steps 3768 and 3769 I0421 11:03:20.275213 47076539613632 model_training_utils.py:505] Train Step: 1670/2100 / loss = 0.7486572265625 I0421 11:03:20.275651 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 3769 and 3770 I0421 11:03:21.356395 47076539613632 model_training_utils.py:505] Train Step: 1671/2100 / loss = 0.7333984375 I0421 11:03:21.356825 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.27 examples/second between steps 3770 and 3771 I0421 11:03:22.438003 47076539613632 model_training_utils.py:505] Train Step: 1672/2100 / loss = 0.8253173828125 I0421 11:03:22.438444 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 3771 and 3772 I0421 11:03:23.523346 47076539613632 model_training_utils.py:505] Train Step: 1673/2100 / loss = 1.0684814453125 I0421 11:03:23.523784 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 3772 and 3773 I0421 11:03:24.607521 47076539613632 model_training_utils.py:505] Train Step: 1674/2100 / loss = 0.8079833984375 I0421 11:03:24.607962 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 3773 and 3774 I0421 11:03:25.692753 47076539613632 model_training_utils.py:505] Train Step: 1675/2100 / loss = 0.917236328125 I0421 11:03:25.693175 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 3774 and 3775 I0421 11:03:26.779956 47076539613632 model_training_utils.py:505] Train Step: 1676/2100 / loss = 1.41162109375 I0421 11:03:26.780380 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.62 examples/second between steps 3775 and 3776 I0421 11:03:27.864591 47076539613632 model_training_utils.py:505] Train Step: 1677/2100 / loss = 2.009765625 I0421 11:03:27.865015 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 3776 and 3777 I0421 11:03:28.951037 47076539613632 model_training_utils.py:505] Train Step: 1678/2100 / loss = 1.534423828125 I0421 11:03:28.951473 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 3777 and 3778 I0421 11:03:30.036000 47076539613632 model_training_utils.py:505] Train Step: 1679/2100 / loss = 1.0703125 I0421 11:03:30.036432 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 3778 and 3779 I0421 11:03:31.121229 47076539613632 model_training_utils.py:505] Train Step: 1680/2100 / loss = 0.72265625 I0421 11:03:31.121664 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 3779 and 3780 I0421 11:03:32.206182 47076539613632 model_training_utils.py:505] Train Step: 1681/2100 / loss = 0.6400146484375 I0421 11:03:32.206611 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 3780 and 3781 I0421 11:03:33.290997 47076539613632 model_training_utils.py:505] Train Step: 1682/2100 / loss = 0.694091796875 I0421 11:03:33.291429 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 3781 and 3782 I0421 11:03:34.375036 47076539613632 model_training_utils.py:505] Train Step: 1683/2100 / loss = 0.933837890625 I0421 11:03:34.375477 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 3782 and 3783 I0421 11:03:35.459881 47076539613632 model_training_utils.py:505] Train Step: 1684/2100 / loss = 1.043212890625 I0421 11:03:35.460312 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 3783 and 3784 I0421 11:03:36.545391 47076539613632 model_training_utils.py:505] Train Step: 1685/2100 / loss = 1.334716796875 I0421 11:03:36.545819 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 3784 and 3785 I0421 11:03:37.630089 47076539613632 model_training_utils.py:505] Train Step: 1686/2100 / loss = 1.6787109375 I0421 11:03:37.630527 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 3785 and 3786 I0421 11:03:38.713257 47076539613632 model_training_utils.py:505] Train Step: 1687/2100 / loss = 1.40869140625 I0421 11:03:38.713700 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.07 examples/second between steps 3786 and 3787 I0421 11:03:39.798079 47076539613632 model_training_utils.py:505] Train Step: 1688/2100 / loss = 1.000732421875 I0421 11:03:39.798516 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 3787 and 3788 I0421 11:03:40.882076 47076539613632 model_training_utils.py:505] Train Step: 1689/2100 / loss = 2.242431640625 I0421 11:03:40.882516 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 3788 and 3789 I0421 11:03:41.966592 47076539613632 model_training_utils.py:505] Train Step: 1690/2100 / loss = 2.2265625 I0421 11:03:41.967020 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 3789 and 3790 I0421 11:03:43.048331 47076539613632 model_training_utils.py:505] Train Step: 1691/2100 / loss = 1.468994140625 I0421 11:03:43.048758 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 3790 and 3791 I0421 11:03:44.129219 47076539613632 model_training_utils.py:505] Train Step: 1692/2100 / loss = 1.048583984375 I0421 11:03:44.129671 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.33 examples/second between steps 3791 and 3792 I0421 11:03:45.210409 47076539613632 model_training_utils.py:505] Train Step: 1693/2100 / loss = 0.76904296875 I0421 11:03:45.210834 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.32 examples/second between steps 3792 and 3793 I0421 11:03:46.293045 47076539613632 model_training_utils.py:505] Train Step: 1694/2100 / loss = 0.8448486328125 I0421 11:03:46.293478 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 3793 and 3794 I0421 11:03:47.373048 47076539613632 model_training_utils.py:505] Train Step: 1695/2100 / loss = 0.738037109375 I0421 11:03:47.373480 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.43 examples/second between steps 3794 and 3795 I0421 11:03:48.449594 47076539613632 model_training_utils.py:505] Train Step: 1696/2100 / loss = 0.911376953125 I0421 11:03:48.450014 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.78 examples/second between steps 3795 and 3796 I0421 11:03:49.532683 47076539613632 model_training_utils.py:505] Train Step: 1697/2100 / loss = 0.861083984375 I0421 11:03:49.533114 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 3796 and 3797 I0421 11:03:50.616101 47076539613632 model_training_utils.py:505] Train Step: 1698/2100 / loss = 0.94482421875 I0421 11:03:50.616534 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 3797 and 3798 I0421 11:03:51.698550 47076539613632 model_training_utils.py:505] Train Step: 1699/2100 / loss = 0.8314208984375 I0421 11:03:51.698971 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 3798 and 3799 I0421 11:03:52.782813 47076539613632 model_training_utils.py:505] Train Step: 1700/2100 / loss = 0.969482421875 I0421 11:03:52.783238 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 3799 and 3800 I0421 11:03:53.866095 47076539613632 model_training_utils.py:505] Train Step: 1701/2100 / loss = 1.116455078125 I0421 11:03:53.866529 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 3800 and 3801 I0421 11:03:54.947920 47076539613632 model_training_utils.py:505] Train Step: 1702/2100 / loss = 0.9039306640625 I0421 11:03:54.948356 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 3801 and 3802 I0421 11:03:56.033125 47076539613632 model_training_utils.py:505] Train Step: 1703/2100 / loss = 0.861083984375 I0421 11:03:56.033553 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 3802 and 3803 I0421 11:03:57.118627 47076539613632 model_training_utils.py:505] Train Step: 1704/2100 / loss = 0.9942626953125 I0421 11:03:57.119055 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 3803 and 3804 I0421 11:03:58.202019 47076539613632 model_training_utils.py:505] Train Step: 1705/2100 / loss = 0.74560546875 I0421 11:03:58.202461 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 3804 and 3805 I0421 11:03:59.286608 47076539613632 model_training_utils.py:505] Train Step: 1706/2100 / loss = 0.6060791015625 I0421 11:03:59.287037 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 3805 and 3806 I0421 11:04:00.367488 47076539613632 model_training_utils.py:505] Train Step: 1707/2100 / loss = 0.6583251953125 I0421 11:04:00.367924 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 3806 and 3807 I0421 11:04:01.450462 47076539613632 model_training_utils.py:505] Train Step: 1708/2100 / loss = 0.6435546875 I0421 11:04:01.450887 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 3807 and 3808 I0421 11:04:02.532881 47076539613632 model_training_utils.py:505] Train Step: 1709/2100 / loss = 0.68084716796875 I0421 11:04:02.533312 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 3808 and 3809 I0421 11:04:03.616086 47076539613632 model_training_utils.py:505] Train Step: 1710/2100 / loss = 0.64208984375 I0421 11:04:03.616518 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 3809 and 3810 I0421 11:04:04.699175 47076539613632 model_training_utils.py:505] Train Step: 1711/2100 / loss = 0.590087890625 I0421 11:04:04.699613 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 3810 and 3811 I0421 11:04:05.782457 47076539613632 model_training_utils.py:505] Train Step: 1712/2100 / loss = 0.40203857421875 I0421 11:04:05.782883 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 3811 and 3812 I0421 11:04:06.864955 47076539613632 model_training_utils.py:505] Train Step: 1713/2100 / loss = 0.550537109375 I0421 11:04:06.865388 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 3812 and 3813 I0421 11:04:07.944384 47076539613632 model_training_utils.py:505] Train Step: 1714/2100 / loss = 0.60296630859375 I0421 11:04:07.944806 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.48 examples/second between steps 3813 and 3814 I0421 11:04:09.031543 47076539613632 model_training_utils.py:505] Train Step: 1715/2100 / loss = 1.6572265625 I0421 11:04:09.031968 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.64 examples/second between steps 3814 and 3815 I0421 11:04:10.117017 47076539613632 model_training_utils.py:505] Train Step: 1716/2100 / loss = 1.0677490234375 I0421 11:04:10.117466 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 3815 and 3816 I0421 11:04:11.198052 47076539613632 model_training_utils.py:505] Train Step: 1717/2100 / loss = 0.906982421875 I0421 11:04:11.198489 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.36 examples/second between steps 3816 and 3817 I0421 11:04:12.282665 47076539613632 model_training_utils.py:505] Train Step: 1718/2100 / loss = 0.63946533203125 I0421 11:04:12.283090 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 3817 and 3818 I0421 11:04:13.366016 47076539613632 model_training_utils.py:505] Train Step: 1719/2100 / loss = 0.51409912109375 I0421 11:04:13.366451 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 3818 and 3819 I0421 11:04:14.448230 47076539613632 model_training_utils.py:505] Train Step: 1720/2100 / loss = 1.6402587890625 I0421 11:04:14.448668 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 3819 and 3820 I0421 11:04:15.531314 47076539613632 model_training_utils.py:505] Train Step: 1721/2100 / loss = 1.91796875 I0421 11:04:15.531739 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.07 examples/second between steps 3820 and 3821 I0421 11:04:16.616837 47076539613632 model_training_utils.py:505] Train Step: 1722/2100 / loss = 0.955810546875 I0421 11:04:16.617263 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 3821 and 3822 I0421 11:04:17.702095 47076539613632 model_training_utils.py:505] Train Step: 1723/2100 / loss = 1.0462646484375 I0421 11:04:17.702527 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 3822 and 3823 I0421 11:04:18.784770 47076539613632 model_training_utils.py:505] Train Step: 1724/2100 / loss = 1.0302734375 I0421 11:04:18.785200 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 3823 and 3824 I0421 11:04:19.867864 47076539613632 model_training_utils.py:505] Train Step: 1725/2100 / loss = 0.751220703125 I0421 11:04:19.868303 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.07 examples/second between steps 3824 and 3825 I0421 11:04:20.951326 47076539613632 model_training_utils.py:505] Train Step: 1726/2100 / loss = 0.8013916015625 I0421 11:04:20.951748 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 3825 and 3826 I0421 11:04:22.035851 47076539613632 model_training_utils.py:505] Train Step: 1727/2100 / loss = 0.8851318359375 I0421 11:04:22.036277 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 3826 and 3827 I0421 11:04:23.123301 47076539613632 model_training_utils.py:505] Train Step: 1728/2100 / loss = 1.200439453125 I0421 11:04:23.123740 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.57 examples/second between steps 3827 and 3828 I0421 11:04:24.207696 47076539613632 model_training_utils.py:505] Train Step: 1729/2100 / loss = 0.7044677734375 I0421 11:04:24.208128 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 3828 and 3829 I0421 11:04:25.290557 47076539613632 model_training_utils.py:505] Train Step: 1730/2100 / loss = 0.639984130859375 I0421 11:04:25.290987 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 3829 and 3830 I0421 11:04:26.375951 47076539613632 model_training_utils.py:505] Train Step: 1731/2100 / loss = 0.843505859375 I0421 11:04:26.376388 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 3830 and 3831 I0421 11:04:27.459451 47076539613632 model_training_utils.py:505] Train Step: 1732/2100 / loss = 1.1055908203125 I0421 11:04:27.459883 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 3831 and 3832 I0421 11:04:28.544592 47076539613632 model_training_utils.py:505] Train Step: 1733/2100 / loss = 1.397216796875 I0421 11:04:28.545011 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 3832 and 3833 I0421 11:04:29.632129 47076539613632 model_training_utils.py:505] Train Step: 1734/2100 / loss = 1.263427734375 I0421 11:04:29.632570 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.57 examples/second between steps 3833 and 3834 I0421 11:04:30.717588 47076539613632 model_training_utils.py:505] Train Step: 1735/2100 / loss = 1.0968017578125 I0421 11:04:30.718016 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 3834 and 3835 I0421 11:04:31.800249 47076539613632 model_training_utils.py:505] Train Step: 1736/2100 / loss = 0.9251708984375 I0421 11:04:31.800682 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 3835 and 3836 I0421 11:04:32.880910 47076539613632 model_training_utils.py:505] Train Step: 1737/2100 / loss = 1.0086669921875 I0421 11:04:32.881345 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.35 examples/second between steps 3836 and 3837 I0421 11:04:33.966488 47076539613632 model_training_utils.py:505] Train Step: 1738/2100 / loss = 0.7049560546875 I0421 11:04:33.966913 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 3837 and 3838 I0421 11:04:35.050962 47076539613632 model_training_utils.py:505] Train Step: 1739/2100 / loss = 0.59381103515625 I0421 11:04:35.051403 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 3838 and 3839 I0421 11:04:36.137094 47076539613632 model_training_utils.py:505] Train Step: 1740/2100 / loss = 0.818359375 I0421 11:04:36.137540 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.71 examples/second between steps 3839 and 3840 I0421 11:04:37.222166 47076539613632 model_training_utils.py:505] Train Step: 1741/2100 / loss = 0.9891357421875 I0421 11:04:37.222602 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.82 examples/second between steps 3840 and 3841 I0421 11:04:38.299886 47076539613632 model_training_utils.py:505] Train Step: 1742/2100 / loss = 1.060302734375 I0421 11:04:38.300322 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.65 examples/second between steps 3841 and 3842 I0421 11:04:39.385161 47076539613632 model_training_utils.py:505] Train Step: 1743/2100 / loss = 1.376953125 I0421 11:04:39.385598 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 3842 and 3843 I0421 11:04:40.471567 47076539613632 model_training_utils.py:505] Train Step: 1744/2100 / loss = 1.757568359375 I0421 11:04:40.471995 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 3843 and 3844 I0421 11:04:41.560921 47076539613632 model_training_utils.py:505] Train Step: 1745/2100 / loss = 1.392822265625 I0421 11:04:41.561351 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.39 examples/second between steps 3844 and 3845 I0421 11:04:42.645910 47076539613632 model_training_utils.py:505] Train Step: 1746/2100 / loss = 1.008056640625 I0421 11:04:42.646345 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 3845 and 3846 I0421 11:04:43.727850 47076539613632 model_training_utils.py:505] Train Step: 1747/2100 / loss = 1.23046875 I0421 11:04:43.728302 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 3846 and 3847 I0421 11:04:44.811581 47076539613632 model_training_utils.py:505] Train Step: 1748/2100 / loss = 0.95721435546875 I0421 11:04:44.812019 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 3847 and 3848 I0421 11:04:45.895220 47076539613632 model_training_utils.py:505] Train Step: 1749/2100 / loss = 1.00830078125 I0421 11:04:45.895650 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 3848 and 3849 I0421 11:04:46.980346 47076539613632 model_training_utils.py:505] Train Step: 1750/2100 / loss = 0.9287109375 I0421 11:04:46.980771 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 3849 and 3850 I0421 11:04:48.065562 47076539613632 model_training_utils.py:505] Train Step: 1751/2100 / loss = 1.0699462890625 I0421 11:04:48.065993 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 3850 and 3851 I0421 11:04:49.149420 47076539613632 model_training_utils.py:505] Train Step: 1752/2100 / loss = 1.039794921875 I0421 11:04:49.149852 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 3851 and 3852 I0421 11:04:50.233353 47076539613632 model_training_utils.py:505] Train Step: 1753/2100 / loss = 1.0596923828125 I0421 11:04:50.233780 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 3852 and 3853 I0421 11:04:51.317146 47076539613632 model_training_utils.py:505] Train Step: 1754/2100 / loss = 0.961669921875 I0421 11:04:51.317573 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 3853 and 3854 I0421 11:04:52.402159 47076539613632 model_training_utils.py:505] Train Step: 1755/2100 / loss = 0.8817138671875 I0421 11:04:52.402594 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 3854 and 3855 I0421 11:04:53.485527 47076539613632 model_training_utils.py:505] Train Step: 1756/2100 / loss = 0.904052734375 I0421 11:04:53.485951 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.07 examples/second between steps 3855 and 3856 I0421 11:04:54.566631 47076539613632 model_training_utils.py:505] Train Step: 1757/2100 / loss = 0.8184814453125 I0421 11:04:54.567059 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.31 examples/second between steps 3856 and 3857 I0421 11:04:55.649698 47076539613632 model_training_utils.py:505] Train Step: 1758/2100 / loss = 0.6905517578125 I0421 11:04:55.650123 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 3857 and 3858 I0421 11:04:56.733181 47076539613632 model_training_utils.py:505] Train Step: 1759/2100 / loss = 0.9476318359375 I0421 11:04:56.733618 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 3858 and 3859 I0421 11:04:57.814695 47076539613632 model_training_utils.py:505] Train Step: 1760/2100 / loss = 2.12158203125 I0421 11:04:57.815117 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 3859 and 3860 I0421 11:04:58.894145 47076539613632 model_training_utils.py:505] Train Step: 1761/2100 / loss = 1.36767578125 I0421 11:04:58.894575 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.48 examples/second between steps 3860 and 3861 I0421 11:04:59.974485 47076539613632 model_training_utils.py:505] Train Step: 1762/2100 / loss = 1.04150390625 I0421 11:04:59.974916 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.38 examples/second between steps 3861 and 3862 I0421 11:05:01.054909 47076539613632 model_training_utils.py:505] Train Step: 1763/2100 / loss = 1.191650390625 I0421 11:05:01.055352 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.37 examples/second between steps 3862 and 3863 I0421 11:05:02.136842 47076539613632 model_training_utils.py:505] Train Step: 1764/2100 / loss = 0.9566650390625 I0421 11:05:02.137266 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 3863 and 3864 I0421 11:05:03.219763 47076539613632 model_training_utils.py:505] Train Step: 1765/2100 / loss = 1.520263671875 I0421 11:05:03.220184 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 3864 and 3865 I0421 11:05:04.302915 47076539613632 model_training_utils.py:505] Train Step: 1766/2100 / loss = 1.6220703125 I0421 11:05:04.303347 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 3865 and 3866 I0421 11:05:05.386198 47076539613632 model_training_utils.py:505] Train Step: 1767/2100 / loss = 1.318603515625 I0421 11:05:05.386633 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 3866 and 3867 I0421 11:05:06.467348 47076539613632 model_training_utils.py:505] Train Step: 1768/2100 / loss = 0.764404296875 I0421 11:05:06.467772 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.29 examples/second between steps 3867 and 3868 I0421 11:05:07.548090 47076539613632 model_training_utils.py:505] Train Step: 1769/2100 / loss = 0.763916015625 I0421 11:05:07.548535 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.36 examples/second between steps 3868 and 3869 I0421 11:05:08.629791 47076539613632 model_training_utils.py:505] Train Step: 1770/2100 / loss = 1.3099365234375 I0421 11:05:08.630218 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 3869 and 3870 I0421 11:05:09.710860 47076539613632 model_training_utils.py:505] Train Step: 1771/2100 / loss = 1.9794921875 I0421 11:05:09.711323 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.34 examples/second between steps 3870 and 3871 I0421 11:05:10.792884 47076539613632 model_training_utils.py:505] Train Step: 1772/2100 / loss = 0.998291015625 I0421 11:05:10.793319 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 3871 and 3872 I0421 11:05:11.873722 47076539613632 model_training_utils.py:505] Train Step: 1773/2100 / loss = 0.71630859375 I0421 11:05:11.874159 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.31 examples/second between steps 3872 and 3873 I0421 11:05:12.955690 47076539613632 model_training_utils.py:505] Train Step: 1774/2100 / loss = 1.0521240234375 I0421 11:05:12.956111 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 3873 and 3874 I0421 11:05:14.040569 47076539613632 model_training_utils.py:505] Train Step: 1775/2100 / loss = 0.7998046875 I0421 11:05:14.040996 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 3874 and 3875 I0421 11:05:15.125716 47076539613632 model_training_utils.py:505] Train Step: 1776/2100 / loss = 0.73388671875 I0421 11:05:15.126147 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 3875 and 3876 I0421 11:05:16.203854 47076539613632 model_training_utils.py:505] Train Step: 1777/2100 / loss = 0.9169921875 I0421 11:05:16.204292 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.61 examples/second between steps 3876 and 3877 I0421 11:05:17.287431 47076539613632 model_training_utils.py:505] Train Step: 1778/2100 / loss = 1.1474609375 I0421 11:05:17.287862 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 3877 and 3878 I0421 11:05:18.370485 47076539613632 model_training_utils.py:505] Train Step: 1779/2100 / loss = 1.87646484375 I0421 11:05:18.370922 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 3878 and 3879 I0421 11:05:19.457390 47076539613632 model_training_utils.py:505] Train Step: 1780/2100 / loss = 2.462890625 I0421 11:05:19.457822 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.64 examples/second between steps 3879 and 3880 I0421 11:05:20.543172 47076539613632 model_training_utils.py:505] Train Step: 1781/2100 / loss = 2.630859375 I0421 11:05:20.543609 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 3880 and 3881 I0421 11:05:21.622077 47076539613632 model_training_utils.py:505] Train Step: 1782/2100 / loss = 1.4278564453125 I0421 11:05:21.622516 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.54 examples/second between steps 3881 and 3882 I0421 11:05:22.704986 47076539613632 model_training_utils.py:505] Train Step: 1783/2100 / loss = 1.20458984375 I0421 11:05:22.705420 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 3882 and 3883 I0421 11:05:23.789163 47076539613632 model_training_utils.py:505] Train Step: 1784/2100 / loss = 1.2744140625 I0421 11:05:23.789596 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 3883 and 3884 I0421 11:05:24.872061 47076539613632 model_training_utils.py:505] Train Step: 1785/2100 / loss = 0.883544921875 I0421 11:05:24.872497 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 3884 and 3885 I0421 11:05:25.954165 47076539613632 model_training_utils.py:505] Train Step: 1786/2100 / loss = 0.81982421875 I0421 11:05:25.954602 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 3885 and 3886 I0421 11:05:27.033933 47076539613632 model_training_utils.py:505] Train Step: 1787/2100 / loss = 0.7591552734375 I0421 11:05:27.034375 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.39 examples/second between steps 3886 and 3887 I0421 11:05:28.116427 47076539613632 model_training_utils.py:505] Train Step: 1788/2100 / loss = 0.772216796875 I0421 11:05:28.116851 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 3887 and 3888 I0421 11:05:29.199769 47076539613632 model_training_utils.py:505] Train Step: 1789/2100 / loss = 1.4951171875 I0421 11:05:29.200204 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 3888 and 3889 I0421 11:05:30.286278 47076539613632 model_training_utils.py:505] Train Step: 1790/2100 / loss = 1.94677734375 I0421 11:05:30.286721 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.67 examples/second between steps 3889 and 3890 I0421 11:05:31.369552 47076539613632 model_training_utils.py:505] Train Step: 1791/2100 / loss = 1.6533203125 I0421 11:05:31.369983 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 3890 and 3891 I0421 11:05:32.453461 47076539613632 model_training_utils.py:505] Train Step: 1792/2100 / loss = 1.2333984375 I0421 11:05:32.453906 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.96 examples/second between steps 3891 and 3892 I0421 11:05:33.532789 47076539613632 model_training_utils.py:505] Train Step: 1793/2100 / loss = 0.6387939453125 I0421 11:05:33.533217 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.46 examples/second between steps 3892 and 3893 I0421 11:05:34.615842 47076539613632 model_training_utils.py:505] Train Step: 1794/2100 / loss = 0.5743408203125 I0421 11:05:34.616273 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 3893 and 3894 I0421 11:05:35.700790 47076539613632 model_training_utils.py:505] Train Step: 1795/2100 / loss = 0.95751953125 I0421 11:05:35.701166 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 3894 and 3895 I0421 11:05:36.787060 47076539613632 model_training_utils.py:505] Train Step: 1796/2100 / loss = 0.9722900390625 I0421 11:05:36.787459 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 3895 and 3896 I0421 11:05:37.867507 47076539613632 model_training_utils.py:505] Train Step: 1797/2100 / loss = 1.067626953125 I0421 11:05:37.867889 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.38 examples/second between steps 3896 and 3897 I0421 11:05:38.947132 47076539613632 model_training_utils.py:505] Train Step: 1798/2100 / loss = 0.97509765625 I0421 11:05:38.947531 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.50 examples/second between steps 3897 and 3898 I0421 11:05:40.032845 47076539613632 model_training_utils.py:505] Train Step: 1799/2100 / loss = 0.857421875 I0421 11:05:40.033232 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 3898 and 3899 I0421 11:05:41.118586 47076539613632 model_training_utils.py:505] Train Step: 1800/2100 / loss = 0.6942138671875 I0421 11:05:41.118969 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 3899 and 3900 I0421 11:05:42.203112 47076539613632 model_training_utils.py:505] Train Step: 1801/2100 / loss = 0.69049072265625 I0421 11:05:42.203496 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 3900 and 3901 I0421 11:05:43.284675 47076539613632 model_training_utils.py:505] Train Step: 1802/2100 / loss = 0.774169921875 I0421 11:05:43.285057 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 3901 and 3902 I0421 11:05:44.368676 47076539613632 model_training_utils.py:505] Train Step: 1803/2100 / loss = 0.9024658203125 I0421 11:05:44.369068 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 3902 and 3903 I0421 11:05:45.451138 47076539613632 model_training_utils.py:505] Train Step: 1804/2100 / loss = 0.9337158203125 I0421 11:05:45.451530 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 3903 and 3904 I0421 11:05:46.536096 47076539613632 model_training_utils.py:505] Train Step: 1805/2100 / loss = 0.9898681640625 I0421 11:05:46.536499 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 3904 and 3905 I0421 11:05:47.619272 47076539613632 model_training_utils.py:505] Train Step: 1806/2100 / loss = 1.263916015625 I0421 11:05:47.619658 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 3905 and 3906 I0421 11:05:48.702426 47076539613632 model_training_utils.py:505] Train Step: 1807/2100 / loss = 1.319580078125 I0421 11:05:48.702808 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 3906 and 3907 I0421 11:05:49.785095 47076539613632 model_training_utils.py:505] Train Step: 1808/2100 / loss = 0.84814453125 I0421 11:05:49.785485 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 3907 and 3908 I0421 11:05:50.871025 47076539613632 model_training_utils.py:505] Train Step: 1809/2100 / loss = 0.800537109375 I0421 11:05:50.871416 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.77 examples/second between steps 3908 and 3909 I0421 11:05:51.953616 47076539613632 model_training_utils.py:505] Train Step: 1810/2100 / loss = 0.8670654296875 I0421 11:05:51.954001 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 3909 and 3910 I0421 11:05:53.035529 47076539613632 model_training_utils.py:505] Train Step: 1811/2100 / loss = 1.260986328125 I0421 11:05:53.035905 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 3910 and 3911 I0421 11:05:54.117899 47076539613632 model_training_utils.py:505] Train Step: 1812/2100 / loss = 1.4443359375 I0421 11:05:54.118295 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 3911 and 3912 I0421 11:05:55.203085 47076539613632 model_training_utils.py:505] Train Step: 1813/2100 / loss = 1.0185546875 I0421 11:05:55.203475 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 3912 and 3913 I0421 11:05:56.285157 47076539613632 model_training_utils.py:505] Train Step: 1814/2100 / loss = 1.0352783203125 I0421 11:05:56.285554 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 3913 and 3914 I0421 11:05:57.369049 47076539613632 model_training_utils.py:505] Train Step: 1815/2100 / loss = 0.886474609375 I0421 11:05:57.369438 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 3914 and 3915 I0421 11:05:58.451062 47076539613632 model_training_utils.py:505] Train Step: 1816/2100 / loss = 0.6058349609375 I0421 11:05:58.451452 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 3915 and 3916 I0421 11:05:59.532935 47076539613632 model_training_utils.py:505] Train Step: 1817/2100 / loss = 0.84619140625 I0421 11:05:59.533325 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 3916 and 3917 I0421 11:06:00.614482 47076539613632 model_training_utils.py:505] Train Step: 1818/2100 / loss = 1.0762939453125 I0421 11:06:00.614866 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 3917 and 3918 I0421 11:06:01.696983 47076539613632 model_training_utils.py:505] Train Step: 1819/2100 / loss = 1.22119140625 I0421 11:06:01.697373 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 3918 and 3919 I0421 11:06:02.778307 47076539613632 model_training_utils.py:505] Train Step: 1820/2100 / loss = 0.90234375 I0421 11:06:02.778690 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.27 examples/second between steps 3919 and 3920 I0421 11:06:03.861147 47076539613632 model_training_utils.py:505] Train Step: 1821/2100 / loss = 0.91943359375 I0421 11:06:03.861535 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 3920 and 3921 I0421 11:06:04.940239 47076539613632 model_training_utils.py:505] Train Step: 1822/2100 / loss = 1.44677734375 I0421 11:06:04.940633 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.51 examples/second between steps 3921 and 3922 I0421 11:06:06.025058 47076539613632 model_training_utils.py:505] Train Step: 1823/2100 / loss = 1.123779296875 I0421 11:06:06.025454 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 3922 and 3923 I0421 11:06:07.110975 47076539613632 model_training_utils.py:505] Train Step: 1824/2100 / loss = 0.87353515625 I0421 11:06:07.111364 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 3923 and 3924 I0421 11:06:08.199846 47076539613632 model_training_utils.py:505] Train Step: 1825/2100 / loss = 1.69482421875 I0421 11:06:08.200226 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.44 examples/second between steps 3924 and 3925 I0421 11:06:09.284554 47076539613632 model_training_utils.py:505] Train Step: 1826/2100 / loss = 1.6435546875 I0421 11:06:09.284944 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 3925 and 3926 I0421 11:06:10.369580 47076539613632 model_training_utils.py:505] Train Step: 1827/2100 / loss = 1.501953125 I0421 11:06:10.369967 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 3926 and 3927 I0421 11:06:11.453219 47076539613632 model_training_utils.py:505] Train Step: 1828/2100 / loss = 1.88916015625 I0421 11:06:11.453624 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 3927 and 3928 I0421 11:06:12.539495 47076539613632 model_training_utils.py:505] Train Step: 1829/2100 / loss = 1.4432373046875 I0421 11:06:12.539891 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 3928 and 3929 I0421 11:06:13.613253 47076539613632 model_training_utils.py:505] Train Step: 1830/2100 / loss = 1.356201171875 I0421 11:06:13.613632 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 120.13 examples/second between steps 3929 and 3930 I0421 11:06:14.698456 47076539613632 model_training_utils.py:505] Train Step: 1831/2100 / loss = 0.8663330078125 I0421 11:06:14.698845 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 3930 and 3931 I0421 11:06:15.781847 47076539613632 model_training_utils.py:505] Train Step: 1832/2100 / loss = 0.837158203125 I0421 11:06:15.782256 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.07 examples/second between steps 3931 and 3932 I0421 11:06:16.868169 47076539613632 model_training_utils.py:505] Train Step: 1833/2100 / loss = 0.7132568359375 I0421 11:06:16.868572 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 3932 and 3933 I0421 11:06:17.950647 47076539613632 model_training_utils.py:505] Train Step: 1834/2100 / loss = 1.0638427734375 I0421 11:06:17.951043 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 3933 and 3934 I0421 11:06:19.038054 47076539613632 model_training_utils.py:505] Train Step: 1835/2100 / loss = 1.036865234375 I0421 11:06:19.038454 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.62 examples/second between steps 3934 and 3935 I0421 11:06:20.120035 47076539613632 model_training_utils.py:505] Train Step: 1836/2100 / loss = 0.9697265625 I0421 11:06:20.120442 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 3935 and 3936 I0421 11:06:21.201780 47076539613632 model_training_utils.py:505] Train Step: 1837/2100 / loss = 1.20556640625 I0421 11:06:21.202178 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.27 examples/second between steps 3936 and 3937 I0421 11:06:22.282882 47076539613632 model_training_utils.py:505] Train Step: 1838/2100 / loss = 1.039306640625 I0421 11:06:22.283292 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.36 examples/second between steps 3937 and 3938 I0421 11:06:23.368770 47076539613632 model_training_utils.py:505] Train Step: 1839/2100 / loss = 0.814208984375 I0421 11:06:23.369163 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 3938 and 3939 I0421 11:06:24.451058 47076539613632 model_training_utils.py:505] Train Step: 1840/2100 / loss = 0.9454345703125 I0421 11:06:24.451458 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 3939 and 3940 I0421 11:06:25.528910 47076539613632 model_training_utils.py:505] Train Step: 1841/2100 / loss = 1.4775390625 I0421 11:06:25.529339 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.64 examples/second between steps 3940 and 3941 I0421 11:06:26.610461 47076539613632 model_training_utils.py:505] Train Step: 1842/2100 / loss = 2.05126953125 I0421 11:06:26.610847 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.43 examples/second between steps 3941 and 3942 I0421 11:06:27.694077 47076539613632 model_training_utils.py:505] Train Step: 1843/2100 / loss = 1.55517578125 I0421 11:06:27.694491 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 3942 and 3943 I0421 11:06:28.776526 47076539613632 model_training_utils.py:505] Train Step: 1844/2100 / loss = 0.97314453125 I0421 11:06:28.776927 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 3943 and 3944 I0421 11:06:29.859636 47076539613632 model_training_utils.py:505] Train Step: 1845/2100 / loss = 1.2021484375 I0421 11:06:29.860035 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 3944 and 3945 I0421 11:06:30.944298 47076539613632 model_training_utils.py:505] Train Step: 1846/2100 / loss = 1.0482177734375 I0421 11:06:30.944695 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 3945 and 3946 I0421 11:06:32.026543 47076539613632 model_training_utils.py:505] Train Step: 1847/2100 / loss = 0.68603515625 I0421 11:06:32.026936 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 3946 and 3947 I0421 11:06:33.108710 47076539613632 model_training_utils.py:505] Train Step: 1848/2100 / loss = 0.6915283203125 I0421 11:06:33.109108 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 3947 and 3948 I0421 11:06:34.189373 47076539613632 model_training_utils.py:505] Train Step: 1849/2100 / loss = 0.6817626953125 I0421 11:06:34.189777 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.35 examples/second between steps 3948 and 3949 I0421 11:06:35.271010 47076539613632 model_training_utils.py:505] Train Step: 1850/2100 / loss = 1.587890625 I0421 11:06:35.271417 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.27 examples/second between steps 3949 and 3950 I0421 11:06:36.354381 47076539613632 model_training_utils.py:505] Train Step: 1851/2100 / loss = 1.345458984375 I0421 11:06:36.354773 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 3950 and 3951 I0421 11:06:37.435345 47076539613632 model_training_utils.py:505] Train Step: 1852/2100 / loss = 0.9912109375 I0421 11:06:37.435757 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.31 examples/second between steps 3951 and 3952 I0421 11:06:38.517527 47076539613632 model_training_utils.py:505] Train Step: 1853/2100 / loss = 0.9755859375 I0421 11:06:38.517915 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 3952 and 3953 I0421 11:06:39.598258 47076539613632 model_training_utils.py:505] Train Step: 1854/2100 / loss = 1.45556640625 I0421 11:06:39.598666 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.33 examples/second between steps 3953 and 3954 I0421 11:06:40.681200 47076539613632 model_training_utils.py:505] Train Step: 1855/2100 / loss = 1.26806640625 I0421 11:06:40.681605 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 3954 and 3955 I0421 11:06:41.758630 47076539613632 model_training_utils.py:505] Train Step: 1856/2100 / loss = 0.978271484375 I0421 11:06:41.759044 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.74 examples/second between steps 3955 and 3956 I0421 11:06:42.839884 47076539613632 model_training_utils.py:505] Train Step: 1857/2100 / loss = 1.552734375 I0421 11:06:42.840292 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.27 examples/second between steps 3956 and 3957 I0421 11:06:43.924888 47076539613632 model_training_utils.py:505] Train Step: 1858/2100 / loss = 1.8896484375 I0421 11:06:43.925302 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.88 examples/second between steps 3957 and 3958 I0421 11:06:45.007251 47076539613632 model_training_utils.py:505] Train Step: 1859/2100 / loss = 1.5625 I0421 11:06:45.007668 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 3958 and 3959 I0421 11:06:46.093068 47076539613632 model_training_utils.py:505] Train Step: 1860/2100 / loss = 1.755615234375 I0421 11:06:46.093476 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 3959 and 3960 I0421 11:06:47.178734 47076539613632 model_training_utils.py:505] Train Step: 1861/2100 / loss = 1.774169921875 I0421 11:06:47.179134 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 3960 and 3961 I0421 11:06:48.259132 47076539613632 model_training_utils.py:505] Train Step: 1862/2100 / loss = 1.1494140625 I0421 11:06:48.259538 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.40 examples/second between steps 3961 and 3962 I0421 11:06:49.341525 47076539613632 model_training_utils.py:505] Train Step: 1863/2100 / loss = 0.99072265625 I0421 11:06:49.341926 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 3962 and 3963 I0421 11:06:50.424321 47076539613632 model_training_utils.py:505] Train Step: 1864/2100 / loss = 0.895751953125 I0421 11:06:50.424713 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 3963 and 3964 I0421 11:06:51.508253 47076539613632 model_training_utils.py:505] Train Step: 1865/2100 / loss = 0.8896484375 I0421 11:06:51.508678 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 3964 and 3965 I0421 11:06:52.596349 47076539613632 model_training_utils.py:505] Train Step: 1866/2100 / loss = 1.1561279296875 I0421 11:06:52.596743 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 3965 and 3966 I0421 11:06:53.685327 47076539613632 model_training_utils.py:505] Train Step: 1867/2100 / loss = 0.9893798828125 I0421 11:06:53.685726 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.46 examples/second between steps 3966 and 3967 I0421 11:06:54.771000 47076539613632 model_training_utils.py:505] Train Step: 1868/2100 / loss = 1.1143798828125 I0421 11:06:54.771413 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 3967 and 3968 I0421 11:06:55.854388 47076539613632 model_training_utils.py:505] Train Step: 1869/2100 / loss = 1.0906982421875 I0421 11:06:55.854777 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 3968 and 3969 I0421 11:06:56.939941 47076539613632 model_training_utils.py:505] Train Step: 1870/2100 / loss = 1.348388671875 I0421 11:06:56.940337 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 3969 and 3970 I0421 11:06:58.021109 47076539613632 model_training_utils.py:505] Train Step: 1871/2100 / loss = 1.263671875 I0421 11:06:58.021516 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.30 examples/second between steps 3970 and 3971 I0421 11:06:59.105617 47076539613632 model_training_utils.py:505] Train Step: 1872/2100 / loss = 0.8603515625 I0421 11:06:59.106011 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 3971 and 3972 I0421 11:07:00.190436 47076539613632 model_training_utils.py:505] Train Step: 1873/2100 / loss = 0.9302978515625 I0421 11:07:00.190833 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 3972 and 3973 I0421 11:07:01.276191 47076539613632 model_training_utils.py:505] Train Step: 1874/2100 / loss = 0.8890380859375 I0421 11:07:01.276594 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.78 examples/second between steps 3973 and 3974 I0421 11:07:02.359276 47076539613632 model_training_utils.py:505] Train Step: 1875/2100 / loss = 0.9837646484375 I0421 11:07:02.359677 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 3974 and 3975 I0421 11:07:03.439321 47076539613632 model_training_utils.py:505] Train Step: 1876/2100 / loss = 1.05517578125 I0421 11:07:03.439720 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.40 examples/second between steps 3975 and 3976 I0421 11:07:04.522618 47076539613632 model_training_utils.py:505] Train Step: 1877/2100 / loss = 0.80615234375 I0421 11:07:04.523023 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 3976 and 3977 I0421 11:07:05.604279 47076539613632 model_training_utils.py:505] Train Step: 1878/2100 / loss = 0.76611328125 I0421 11:07:05.604681 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 3977 and 3978 I0421 11:07:06.685694 47076539613632 model_training_utils.py:505] Train Step: 1879/2100 / loss = 0.83837890625 I0421 11:07:06.686094 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 3978 and 3979 I0421 11:07:07.764863 47076539613632 model_training_utils.py:505] Train Step: 1880/2100 / loss = 1.1494140625 I0421 11:07:07.765262 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.51 examples/second between steps 3979 and 3980 I0421 11:07:08.848850 47076539613632 model_training_utils.py:505] Train Step: 1881/2100 / loss = 0.85205078125 I0421 11:07:08.849260 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 3980 and 3981 I0421 11:07:09.930059 47076539613632 model_training_utils.py:505] Train Step: 1882/2100 / loss = 0.6856689453125 I0421 11:07:09.930459 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 3981 and 3982 I0421 11:07:11.011666 47076539613632 model_training_utils.py:505] Train Step: 1883/2100 / loss = 0.7530517578125 I0421 11:07:11.012053 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 3982 and 3983 I0421 11:07:12.092985 47076539613632 model_training_utils.py:505] Train Step: 1884/2100 / loss = 0.43865966796875 I0421 11:07:12.093389 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 3983 and 3984 I0421 11:07:13.174043 47076539613632 model_training_utils.py:505] Train Step: 1885/2100 / loss = 0.643798828125 I0421 11:07:13.174449 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.30 examples/second between steps 3984 and 3985 I0421 11:07:14.260214 47076539613632 model_training_utils.py:505] Train Step: 1886/2100 / loss = 0.74871826171875 I0421 11:07:14.260621 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 3985 and 3986 I0421 11:07:15.338789 47076539613632 model_training_utils.py:505] Train Step: 1887/2100 / loss = 0.8310546875 I0421 11:07:15.339183 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.56 examples/second between steps 3986 and 3987 I0421 11:07:16.421920 47076539613632 model_training_utils.py:505] Train Step: 1888/2100 / loss = 0.7784423828125 I0421 11:07:16.422317 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 3987 and 3988 I0421 11:07:17.505500 47076539613632 model_training_utils.py:505] Train Step: 1889/2100 / loss = 0.70947265625 I0421 11:07:17.505914 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 3988 and 3989 I0421 11:07:18.590808 47076539613632 model_training_utils.py:505] Train Step: 1890/2100 / loss = 0.6990966796875 I0421 11:07:18.591207 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 3989 and 3990 I0421 11:07:19.672427 47076539613632 model_training_utils.py:505] Train Step: 1891/2100 / loss = 0.947998046875 I0421 11:07:19.672821 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.25 examples/second between steps 3990 and 3991 I0421 11:07:20.761242 47076539613632 model_training_utils.py:505] Train Step: 1892/2100 / loss = 1.1102294921875 I0421 11:07:20.761640 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.45 examples/second between steps 3991 and 3992 I0421 11:07:21.848194 47076539613632 model_training_utils.py:505] Train Step: 1893/2100 / loss = 1.02294921875 I0421 11:07:21.848598 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.66 examples/second between steps 3992 and 3993 I0421 11:07:22.935407 47076539613632 model_training_utils.py:505] Train Step: 1894/2100 / loss = 0.89013671875 I0421 11:07:22.935804 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 3993 and 3994 I0421 11:07:24.022322 47076539613632 model_training_utils.py:505] Train Step: 1895/2100 / loss = 0.8482666015625 I0421 11:07:24.022731 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.63 examples/second between steps 3994 and 3995 I0421 11:07:25.108580 47076539613632 model_training_utils.py:505] Train Step: 1896/2100 / loss = 1.143310546875 I0421 11:07:25.108985 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 3995 and 3996 I0421 11:07:26.193104 47076539613632 model_training_utils.py:505] Train Step: 1897/2100 / loss = 0.916015625 I0421 11:07:26.193527 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 3996 and 3997 I0421 11:07:27.277459 47076539613632 model_training_utils.py:505] Train Step: 1898/2100 / loss = 0.96826171875 I0421 11:07:27.277854 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 3997 and 3998 I0421 11:07:28.360837 47076539613632 model_training_utils.py:505] Train Step: 1899/2100 / loss = 1.099609375 I0421 11:07:28.361234 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 3998 and 3999 I0421 11:07:29.443178 47076539613632 model_training_utils.py:505] Train Step: 1900/2100 / loss = 1.0037841796875 I0421 11:07:29.443590 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 3999 and 4000 I0421 11:07:30.526531 47076539613632 model_training_utils.py:505] Train Step: 1901/2100 / loss = 1.145263671875 I0421 11:07:30.526929 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 4000 and 4001 I0421 11:07:31.611829 47076539613632 model_training_utils.py:505] Train Step: 1902/2100 / loss = 1.1761474609375 I0421 11:07:31.612245 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 4001 and 4002 I0421 11:07:32.697085 47076539613632 model_training_utils.py:505] Train Step: 1903/2100 / loss = 1.356201171875 I0421 11:07:32.697497 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 4002 and 4003 I0421 11:07:33.780536 47076539613632 model_training_utils.py:505] Train Step: 1904/2100 / loss = 1.470703125 I0421 11:07:33.780958 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 4003 and 4004 I0421 11:07:34.861957 47076539613632 model_training_utils.py:505] Train Step: 1905/2100 / loss = 0.9813232421875 I0421 11:07:34.862377 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.34 examples/second between steps 4004 and 4005 I0421 11:07:35.943010 47076539613632 model_training_utils.py:505] Train Step: 1906/2100 / loss = 1.116943359375 I0421 11:07:35.943407 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.32 examples/second between steps 4005 and 4006 I0421 11:07:37.026740 47076539613632 model_training_utils.py:505] Train Step: 1907/2100 / loss = 1.28515625 I0421 11:07:37.027135 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 4006 and 4007 I0421 11:07:38.109957 47076539613632 model_training_utils.py:505] Train Step: 1908/2100 / loss = 0.931884765625 I0421 11:07:38.110362 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 4007 and 4008 I0421 11:07:39.192309 47076539613632 model_training_utils.py:505] Train Step: 1909/2100 / loss = 0.8951416015625 I0421 11:07:39.192710 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 4008 and 4009 I0421 11:07:40.275611 47076539613632 model_training_utils.py:505] Train Step: 1910/2100 / loss = 0.754150390625 I0421 11:07:40.276013 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 4009 and 4010 I0421 11:07:41.361074 47076539613632 model_training_utils.py:505] Train Step: 1911/2100 / loss = 0.6614990234375 I0421 11:07:41.361475 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 4010 and 4011 I0421 11:07:42.441095 47076539613632 model_training_utils.py:505] Train Step: 1912/2100 / loss = 0.896484375 I0421 11:07:42.441499 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.42 examples/second between steps 4011 and 4012 I0421 11:07:43.524914 47076539613632 model_training_utils.py:505] Train Step: 1913/2100 / loss = 1.090576171875 I0421 11:07:43.525316 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 4012 and 4013 I0421 11:07:44.606862 47076539613632 model_training_utils.py:505] Train Step: 1914/2100 / loss = 1.0054931640625 I0421 11:07:44.607266 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 4013 and 4014 I0421 11:07:45.688991 47076539613632 model_training_utils.py:505] Train Step: 1915/2100 / loss = 0.9498291015625 I0421 11:07:45.689396 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 4014 and 4015 I0421 11:07:46.773836 47076539613632 model_training_utils.py:505] Train Step: 1916/2100 / loss = 1.084228515625 I0421 11:07:46.774230 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 4015 and 4016 I0421 11:07:47.856385 47076539613632 model_training_utils.py:505] Train Step: 1917/2100 / loss = 0.9346923828125 I0421 11:07:47.856786 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.13 examples/second between steps 4016 and 4017 I0421 11:07:48.937007 47076539613632 model_training_utils.py:505] Train Step: 1918/2100 / loss = 0.7066650390625 I0421 11:07:48.937433 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.35 examples/second between steps 4017 and 4018 I0421 11:07:50.019873 47076539613632 model_training_utils.py:505] Train Step: 1919/2100 / loss = 0.632080078125 I0421 11:07:50.020277 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 4018 and 4019 I0421 11:07:51.104743 47076539613632 model_training_utils.py:505] Train Step: 1920/2100 / loss = 0.79248046875 I0421 11:07:51.105142 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.90 examples/second between steps 4019 and 4020 I0421 11:07:52.184886 47076539613632 model_training_utils.py:505] Train Step: 1921/2100 / loss = 0.7789306640625 I0421 11:07:52.185296 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.42 examples/second between steps 4020 and 4021 I0421 11:07:53.268182 47076539613632 model_training_utils.py:505] Train Step: 1922/2100 / loss = 0.748779296875 I0421 11:07:53.268591 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 4021 and 4022 I0421 11:07:54.343884 47076539613632 model_training_utils.py:505] Train Step: 1923/2100 / loss = 0.57684326171875 I0421 11:07:54.344291 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.91 examples/second between steps 4022 and 4023 I0421 11:07:55.426031 47076539613632 model_training_utils.py:505] Train Step: 1924/2100 / loss = 0.6834716796875 I0421 11:07:55.426443 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 4023 and 4024 I0421 11:07:56.509689 47076539613632 model_training_utils.py:505] Train Step: 1925/2100 / loss = 0.6236572265625 I0421 11:07:56.510094 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 4024 and 4025 I0421 11:07:57.591447 47076539613632 model_training_utils.py:505] Train Step: 1926/2100 / loss = 0.62933349609375 I0421 11:07:57.591838 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 4025 and 4026 I0421 11:07:58.677084 47076539613632 model_training_utils.py:505] Train Step: 1927/2100 / loss = 0.789794921875 I0421 11:07:58.677489 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 4026 and 4027 I0421 11:07:59.762028 47076539613632 model_training_utils.py:505] Train Step: 1928/2100 / loss = 1.254150390625 I0421 11:07:59.762437 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 4027 and 4028 I0421 11:08:00.847660 47076539613632 model_training_utils.py:505] Train Step: 1929/2100 / loss = 0.955810546875 I0421 11:08:00.848055 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 4028 and 4029 I0421 11:08:01.931422 47076539613632 model_training_utils.py:505] Train Step: 1930/2100 / loss = 0.865478515625 I0421 11:08:01.931815 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 4029 and 4030 I0421 11:08:03.015518 47076539613632 model_training_utils.py:505] Train Step: 1931/2100 / loss = 0.8475341796875 I0421 11:08:03.015909 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 4030 and 4031 I0421 11:08:04.100064 47076539613632 model_training_utils.py:505] Train Step: 1932/2100 / loss = 1.3173828125 I0421 11:08:04.100471 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 4031 and 4032 I0421 11:08:05.182394 47076539613632 model_training_utils.py:505] Train Step: 1933/2100 / loss = 1.310302734375 I0421 11:08:05.182794 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 4032 and 4033 I0421 11:08:06.263402 47076539613632 model_training_utils.py:505] Train Step: 1934/2100 / loss = 0.8187255859375 I0421 11:08:06.263811 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.31 examples/second between steps 4033 and 4034 I0421 11:08:07.346934 47076539613632 model_training_utils.py:505] Train Step: 1935/2100 / loss = 0.9581298828125 I0421 11:08:07.347338 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 4034 and 4035 I0421 11:08:08.428180 47076539613632 model_training_utils.py:505] Train Step: 1936/2100 / loss = 0.958984375 I0421 11:08:08.428592 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.29 examples/second between steps 4035 and 4036 I0421 11:08:09.511793 47076539613632 model_training_utils.py:505] Train Step: 1937/2100 / loss = 1.1380615234375 I0421 11:08:09.512191 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 4036 and 4037 I0421 11:08:10.594636 47076539613632 model_training_utils.py:505] Train Step: 1938/2100 / loss = 1.0106201171875 I0421 11:08:10.595033 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 4037 and 4038 I0421 11:08:11.676563 47076539613632 model_training_utils.py:505] Train Step: 1939/2100 / loss = 0.81787109375 I0421 11:08:11.676954 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 4038 and 4039 I0421 11:08:12.754552 47076539613632 model_training_utils.py:505] Train Step: 1940/2100 / loss = 0.891845703125 I0421 11:08:12.754952 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.67 examples/second between steps 4039 and 4040 I0421 11:08:13.836232 47076539613632 model_training_utils.py:505] Train Step: 1941/2100 / loss = 0.6226806640625 I0421 11:08:13.836636 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.25 examples/second between steps 4040 and 4041 I0421 11:08:14.918437 47076539613632 model_training_utils.py:505] Train Step: 1942/2100 / loss = 0.9276123046875 I0421 11:08:14.918845 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 4041 and 4042 I0421 11:08:15.999766 47076539613632 model_training_utils.py:505] Train Step: 1943/2100 / loss = 0.7757568359375 I0421 11:08:16.000164 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 4042 and 4043 I0421 11:08:17.082033 47076539613632 model_training_utils.py:505] Train Step: 1944/2100 / loss = 0.87646484375 I0421 11:08:17.082437 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 4043 and 4044 I0421 11:08:18.165462 47076539613632 model_training_utils.py:505] Train Step: 1945/2100 / loss = 0.906005859375 I0421 11:08:18.165863 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 4044 and 4045 I0421 11:08:19.248203 47076539613632 model_training_utils.py:505] Train Step: 1946/2100 / loss = 1.372314453125 I0421 11:08:19.248606 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 4045 and 4046 I0421 11:08:20.328117 47076539613632 model_training_utils.py:505] Train Step: 1947/2100 / loss = 1.2860107421875 I0421 11:08:20.328535 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.42 examples/second between steps 4046 and 4047 I0421 11:08:21.409403 47076539613632 model_training_utils.py:505] Train Step: 1948/2100 / loss = 1.140625 I0421 11:08:21.409799 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.25 examples/second between steps 4047 and 4048 I0421 11:08:22.496897 47076539613632 model_training_utils.py:505] Train Step: 1949/2100 / loss = 1.143798828125 I0421 11:08:22.497302 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.58 examples/second between steps 4048 and 4049 I0421 11:08:23.583342 47076539613632 model_training_utils.py:505] Train Step: 1950/2100 / loss = 0.7818603515625 I0421 11:08:23.583737 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.70 examples/second between steps 4049 and 4050 I0421 11:08:24.669893 47076539613632 model_training_utils.py:505] Train Step: 1951/2100 / loss = 1.1065673828125 I0421 11:08:24.670307 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.70 examples/second between steps 4050 and 4051 I0421 11:08:25.752531 47076539613632 model_training_utils.py:505] Train Step: 1952/2100 / loss = 0.9114990234375 I0421 11:08:25.752933 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 4051 and 4052 I0421 11:08:26.831916 47076539613632 model_training_utils.py:505] Train Step: 1953/2100 / loss = 0.7855224609375 I0421 11:08:26.832320 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.48 examples/second between steps 4052 and 4053 I0421 11:08:27.913300 47076539613632 model_training_utils.py:505] Train Step: 1954/2100 / loss = 1.0911865234375 I0421 11:08:27.913695 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 4053 and 4054 I0421 11:08:28.995052 47076539613632 model_training_utils.py:505] Train Step: 1955/2100 / loss = 1.1884765625 I0421 11:08:28.995463 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 4054 and 4055 I0421 11:08:30.080240 47076539613632 model_training_utils.py:505] Train Step: 1956/2100 / loss = 1.058837890625 I0421 11:08:30.080643 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 4055 and 4056 I0421 11:08:31.158085 47076539613632 model_training_utils.py:505] Train Step: 1957/2100 / loss = 0.9752197265625 I0421 11:08:31.158486 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.68 examples/second between steps 4056 and 4057 I0421 11:08:32.240981 47076539613632 model_training_utils.py:505] Train Step: 1958/2100 / loss = 0.7789306640625 I0421 11:08:32.241388 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 4057 and 4058 I0421 11:08:33.324574 47076539613632 model_training_utils.py:505] Train Step: 1959/2100 / loss = 1.01611328125 I0421 11:08:33.324974 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 4058 and 4059 I0421 11:08:34.410318 47076539613632 model_training_utils.py:505] Train Step: 1960/2100 / loss = 0.923828125 I0421 11:08:34.410717 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.80 examples/second between steps 4059 and 4060 I0421 11:08:35.493688 47076539613632 model_training_utils.py:505] Train Step: 1961/2100 / loss = 1.591064453125 I0421 11:08:35.494083 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 4060 and 4061 I0421 11:08:36.574999 47076539613632 model_training_utils.py:505] Train Step: 1962/2100 / loss = 2.504150390625 I0421 11:08:36.575401 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.30 examples/second between steps 4061 and 4062 I0421 11:08:37.658523 47076539613632 model_training_utils.py:505] Train Step: 1963/2100 / loss = 2.94287109375 I0421 11:08:37.658921 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 4062 and 4063 I0421 11:08:38.740254 47076539613632 model_training_utils.py:505] Train Step: 1964/2100 / loss = 1.910400390625 I0421 11:08:38.740647 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 4063 and 4064 I0421 11:08:39.822058 47076539613632 model_training_utils.py:505] Train Step: 1965/2100 / loss = 1.54931640625 I0421 11:08:39.822464 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 4064 and 4065 I0421 11:08:40.904032 47076539613632 model_training_utils.py:505] Train Step: 1966/2100 / loss = 1.8916015625 I0421 11:08:40.904438 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 4065 and 4066 I0421 11:08:41.986993 47076539613632 model_training_utils.py:505] Train Step: 1967/2100 / loss = 1.122314453125 I0421 11:08:41.987398 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 4066 and 4067 I0421 11:08:43.068499 47076539613632 model_training_utils.py:505] Train Step: 1968/2100 / loss = 0.85107421875 I0421 11:08:43.068906 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.25 examples/second between steps 4067 and 4068 I0421 11:08:44.147656 47076539613632 model_training_utils.py:505] Train Step: 1969/2100 / loss = 0.6722412109375 I0421 11:08:44.148056 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.53 examples/second between steps 4068 and 4069 I0421 11:08:45.245521 47076539613632 model_training_utils.py:505] Train Step: 1970/2100 / loss = 0.7476806640625 I0421 11:08:45.245930 47076539613632 keras_utils.py:133] TimeHistory: 1.09 seconds, 117.52 examples/second between steps 4069 and 4070 I0421 11:08:46.328814 47076539613632 model_training_utils.py:505] Train Step: 1971/2100 / loss = 0.6973876953125 I0421 11:08:46.329212 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.08 examples/second between steps 4070 and 4071 I0421 11:08:47.410238 47076539613632 model_training_utils.py:505] Train Step: 1972/2100 / loss = 0.760986328125 I0421 11:08:47.410640 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.27 examples/second between steps 4071 and 4072 I0421 11:08:48.493453 47076539613632 model_training_utils.py:505] Train Step: 1973/2100 / loss = 1.114990234375 I0421 11:08:48.493863 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.07 examples/second between steps 4072 and 4073 I0421 11:08:49.574587 47076539613632 model_training_utils.py:505] Train Step: 1974/2100 / loss = 1.207275390625 I0421 11:08:49.574991 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.32 examples/second between steps 4073 and 4074 I0421 11:08:50.657203 47076539613632 model_training_utils.py:505] Train Step: 1975/2100 / loss = 1.139892578125 I0421 11:08:50.657605 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 4074 and 4075 I0421 11:08:51.738698 47076539613632 model_training_utils.py:505] Train Step: 1976/2100 / loss = 1.060791015625 I0421 11:08:51.739094 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 4075 and 4076 I0421 11:08:52.822189 47076539613632 model_training_utils.py:505] Train Step: 1977/2100 / loss = 1.236083984375 I0421 11:08:52.822588 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 4076 and 4077 I0421 11:08:53.908229 47076539613632 model_training_utils.py:505] Train Step: 1978/2100 / loss = 2.52392578125 I0421 11:08:53.908637 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.76 examples/second between steps 4077 and 4078 I0421 11:08:54.995747 47076539613632 model_training_utils.py:505] Train Step: 1979/2100 / loss = 1.609130859375 I0421 11:08:54.996145 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.59 examples/second between steps 4078 and 4079 I0421 11:08:56.078734 47076539613632 model_training_utils.py:505] Train Step: 1980/2100 / loss = 0.79364013671875 I0421 11:08:56.079126 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 4079 and 4080 I0421 11:08:57.160768 47076539613632 model_training_utils.py:505] Train Step: 1981/2100 / loss = 0.8441162109375 I0421 11:08:57.161164 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 4080 and 4081 I0421 11:08:58.243497 47076539613632 model_training_utils.py:505] Train Step: 1982/2100 / loss = 0.78564453125 I0421 11:08:58.243899 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 4081 and 4082 I0421 11:08:59.325234 47076539613632 model_training_utils.py:505] Train Step: 1983/2100 / loss = 0.810791015625 I0421 11:08:59.325643 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 4082 and 4083 I0421 11:09:00.409368 47076539613632 model_training_utils.py:505] Train Step: 1984/2100 / loss = 0.772216796875 I0421 11:09:00.409760 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 4083 and 4084 I0421 11:09:01.494008 47076539613632 model_training_utils.py:505] Train Step: 1985/2100 / loss = 1.15576171875 I0421 11:09:01.494415 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.04 examples/second between steps 4084 and 4085 I0421 11:09:02.580603 47076539613632 model_training_utils.py:505] Train Step: 1986/2100 / loss = 1.385986328125 I0421 11:09:02.580996 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 4085 and 4086 I0421 11:09:03.666307 47076539613632 model_training_utils.py:505] Train Step: 1987/2100 / loss = 1.449462890625 I0421 11:09:03.666705 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.81 examples/second between steps 4086 and 4087 I0421 11:09:04.750313 47076539613632 model_training_utils.py:505] Train Step: 1988/2100 / loss = 1.04052734375 I0421 11:09:04.750716 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.98 examples/second between steps 4087 and 4088 I0421 11:09:05.833278 47076539613632 model_training_utils.py:505] Train Step: 1989/2100 / loss = 0.82275390625 I0421 11:09:05.833678 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 4088 and 4089 I0421 11:09:06.919090 47076539613632 model_training_utils.py:505] Train Step: 1990/2100 / loss = 1.268310546875 I0421 11:09:06.919501 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 4089 and 4090 I0421 11:09:08.000754 47076539613632 model_training_utils.py:505] Train Step: 1991/2100 / loss = 2.58984375 I0421 11:09:08.001147 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 4090 and 4091 I0421 11:09:09.086204 47076539613632 model_training_utils.py:505] Train Step: 1992/2100 / loss = 2.494140625 I0421 11:09:09.086614 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.84 examples/second between steps 4091 and 4092 I0421 11:09:10.170823 47076539613632 model_training_utils.py:505] Train Step: 1993/2100 / loss = 1.392333984375 I0421 11:09:10.171218 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 4092 and 4093 I0421 11:09:11.257383 47076539613632 model_training_utils.py:505] Train Step: 1994/2100 / loss = 1.194580078125 I0421 11:09:11.257812 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.72 examples/second between steps 4093 and 4094 I0421 11:09:12.343500 47076539613632 model_training_utils.py:505] Train Step: 1995/2100 / loss = 0.868408203125 I0421 11:09:12.343913 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 4094 and 4095 I0421 11:09:13.423915 47076539613632 model_training_utils.py:505] Train Step: 1996/2100 / loss = 0.9560546875 I0421 11:09:13.424326 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.37 examples/second between steps 4095 and 4096 I0421 11:09:14.507614 47076539613632 model_training_utils.py:505] Train Step: 1997/2100 / loss = 1.123779296875 I0421 11:09:14.508014 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.00 examples/second between steps 4096 and 4097 I0421 11:09:15.596511 47076539613632 model_training_utils.py:505] Train Step: 1998/2100 / loss = 1.161865234375 I0421 11:09:15.596913 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.41 examples/second between steps 4097 and 4098 I0421 11:09:16.680074 47076539613632 model_training_utils.py:505] Train Step: 1999/2100 / loss = 1.322265625 I0421 11:09:16.680480 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 4098 and 4099 I0421 11:09:17.763973 47076539613632 model_training_utils.py:505] Train Step: 2000/2100 / loss = 1.28369140625 I0421 11:09:17.764382 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 4099 and 4100 I0421 11:09:20.042635 47076539613632 model_training_utils.py:505] Train Step: 2001/2100 / loss = 1.3564453125 I0421 11:09:20.043022 47076539613632 keras_utils.py:133] TimeHistory: 2.27 seconds, 56.37 examples/second between steps 4100 and 4101 I0421 11:09:21.155511 47076539613632 model_training_utils.py:505] Train Step: 2002/2100 / loss = 1.3330078125 I0421 11:09:21.155908 47076539613632 keras_utils.py:133] TimeHistory: 1.10 seconds, 115.86 examples/second between steps 4101 and 4102 I0421 11:09:22.237655 47076539613632 model_training_utils.py:505] Train Step: 2003/2100 / loss = 1.686279296875 I0421 11:09:22.238051 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.22 examples/second between steps 4102 and 4103 I0421 11:09:23.321994 47076539613632 model_training_utils.py:505] Train Step: 2004/2100 / loss = 1.443603515625 I0421 11:09:23.322404 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 4103 and 4104 I0421 11:09:24.403226 47076539613632 model_training_utils.py:505] Train Step: 2005/2100 / loss = 0.9130859375 I0421 11:09:24.403645 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.31 examples/second between steps 4104 and 4105 I0421 11:09:25.486567 47076539613632 model_training_utils.py:505] Train Step: 2006/2100 / loss = 1.0545654296875 I0421 11:09:25.486973 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.09 examples/second between steps 4105 and 4106 I0421 11:09:26.568612 47076539613632 model_training_utils.py:505] Train Step: 2007/2100 / loss = 0.7109375 I0421 11:09:26.569006 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 4106 and 4107 I0421 11:09:27.651691 47076539613632 model_training_utils.py:505] Train Step: 2008/2100 / loss = 0.868896484375 I0421 11:09:27.652094 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 4107 and 4108 I0421 11:09:28.730808 47076539613632 model_training_utils.py:505] Train Step: 2009/2100 / loss = 0.7899169921875 I0421 11:09:28.731206 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.52 examples/second between steps 4108 and 4109 I0421 11:09:29.812163 47076539613632 model_training_utils.py:505] Train Step: 2010/2100 / loss = 0.6884765625 I0421 11:09:29.812570 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.26 examples/second between steps 4109 and 4110 I0421 11:09:30.894645 47076539613632 model_training_utils.py:505] Train Step: 2011/2100 / loss = 1.0849609375 I0421 11:09:30.895039 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 4110 and 4111 I0421 11:09:31.978685 47076539613632 model_training_utils.py:505] Train Step: 2012/2100 / loss = 1.076904296875 I0421 11:09:31.979086 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 4111 and 4112 I0421 11:09:33.061550 47076539613632 model_training_utils.py:505] Train Step: 2013/2100 / loss = 0.82373046875 I0421 11:09:33.061949 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 4112 and 4113 I0421 11:09:34.145255 47076539613632 model_training_utils.py:505] Train Step: 2014/2100 / loss = 0.82373046875 I0421 11:09:34.145664 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 4113 and 4114 I0421 11:09:35.227057 47076539613632 model_training_utils.py:505] Train Step: 2015/2100 / loss = 0.779541015625 I0421 11:09:35.227467 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 4114 and 4115 I0421 11:09:36.310437 47076539613632 model_training_utils.py:505] Train Step: 2016/2100 / loss = 0.7589111328125 I0421 11:09:36.310838 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 4115 and 4116 I0421 11:09:37.393140 47076539613632 model_training_utils.py:505] Train Step: 2017/2100 / loss = 1.141357421875 I0421 11:09:37.393544 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 4116 and 4117 I0421 11:09:38.475077 47076539613632 model_training_utils.py:505] Train Step: 2018/2100 / loss = 1.44384765625 I0421 11:09:38.475487 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.20 examples/second between steps 4117 and 4118 I0421 11:09:39.556672 47076539613632 model_training_utils.py:505] Train Step: 2019/2100 / loss = 1.188232421875 I0421 11:09:39.557084 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 4118 and 4119 I0421 11:09:40.639932 47076539613632 model_training_utils.py:505] Train Step: 2020/2100 / loss = 1.643798828125 I0421 11:09:40.640342 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 4119 and 4120 I0421 11:09:41.724256 47076539613632 model_training_utils.py:505] Train Step: 2021/2100 / loss = 1.966552734375 I0421 11:09:41.724660 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 4120 and 4121 I0421 11:09:42.807525 47076539613632 model_training_utils.py:505] Train Step: 2022/2100 / loss = 2.724609375 I0421 11:09:42.807924 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 4121 and 4122 I0421 11:09:43.888807 47076539613632 model_training_utils.py:505] Train Step: 2023/2100 / loss = 1.811767578125 I0421 11:09:43.889204 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.30 examples/second between steps 4122 and 4123 I0421 11:09:44.973999 47076539613632 model_training_utils.py:505] Train Step: 2024/2100 / loss = 0.9080810546875 I0421 11:09:44.974405 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.85 examples/second between steps 4123 and 4124 I0421 11:09:46.058585 47076539613632 model_training_utils.py:505] Train Step: 2025/2100 / loss = 0.939453125 I0421 11:09:46.058981 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 4124 and 4125 I0421 11:09:47.142748 47076539613632 model_training_utils.py:505] Train Step: 2026/2100 / loss = 0.96826171875 I0421 11:09:47.143176 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 4125 and 4126 I0421 11:09:48.224505 47076539613632 model_training_utils.py:505] Train Step: 2027/2100 / loss = 0.8663330078125 I0421 11:09:48.224934 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 4126 and 4127 I0421 11:09:49.306618 47076539613632 model_training_utils.py:505] Train Step: 2028/2100 / loss = 0.734619140625 I0421 11:09:49.307049 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.23 examples/second between steps 4127 and 4128 I0421 11:09:50.389726 47076539613632 model_training_utils.py:505] Train Step: 2029/2100 / loss = 0.68212890625 I0421 11:09:50.390155 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.11 examples/second between steps 4128 and 4129 I0421 11:09:51.474302 47076539613632 model_training_utils.py:505] Train Step: 2030/2100 / loss = 0.674072265625 I0421 11:09:51.474734 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 4129 and 4130 I0421 11:09:52.560062 47076539613632 model_training_utils.py:505] Train Step: 2031/2100 / loss = 0.7110595703125 I0421 11:09:52.560494 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.79 examples/second between steps 4130 and 4131 I0421 11:09:53.643651 47076539613632 model_training_utils.py:505] Train Step: 2032/2100 / loss = 0.74755859375 I0421 11:09:53.644080 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 4131 and 4132 I0421 11:09:54.726447 47076539613632 model_training_utils.py:505] Train Step: 2033/2100 / loss = 0.922119140625 I0421 11:09:54.726880 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 4132 and 4133 I0421 11:09:55.809819 47076539613632 model_training_utils.py:505] Train Step: 2034/2100 / loss = 0.7923583984375 I0421 11:09:55.810248 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 4133 and 4134 I0421 11:09:56.892173 47076539613632 model_training_utils.py:505] Train Step: 2035/2100 / loss = 0.8277587890625 I0421 11:09:56.892615 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.19 examples/second between steps 4134 and 4135 I0421 11:09:57.977071 47076539613632 model_training_utils.py:505] Train Step: 2036/2100 / loss = 0.73876953125 I0421 11:09:57.977525 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 4135 and 4136 I0421 11:09:59.060828 47076539613632 model_training_utils.py:505] Train Step: 2037/2100 / loss = 1.1414794921875 I0421 11:09:59.061254 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.03 examples/second between steps 4136 and 4137 I0421 11:10:00.141387 47076539613632 model_training_utils.py:505] Train Step: 2038/2100 / loss = 0.8997802734375 I0421 11:10:00.141823 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.37 examples/second between steps 4137 and 4138 I0421 11:10:01.227880 47076539613632 model_training_utils.py:505] Train Step: 2039/2100 / loss = 0.8743896484375 I0421 11:10:01.228310 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.74 examples/second between steps 4138 and 4139 I0421 11:10:02.306494 47076539613632 model_training_utils.py:505] Train Step: 2040/2100 / loss = 0.921142578125 I0421 11:10:02.306927 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.62 examples/second between steps 4139 and 4140 I0421 11:10:03.389000 47076539613632 model_training_utils.py:505] Train Step: 2041/2100 / loss = 0.965576171875 I0421 11:10:03.389434 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 4140 and 4141 I0421 11:10:04.474348 47076539613632 model_training_utils.py:505] Train Step: 2042/2100 / loss = 1.0679931640625 I0421 11:10:04.474779 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.86 examples/second between steps 4141 and 4142 I0421 11:10:05.559686 47076539613632 model_training_utils.py:505] Train Step: 2043/2100 / loss = 0.9359130859375 I0421 11:10:05.560117 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 4142 and 4143 I0421 11:10:06.647274 47076539613632 model_training_utils.py:505] Train Step: 2044/2100 / loss = 0.7901611328125 I0421 11:10:06.647707 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.60 examples/second between steps 4143 and 4144 I0421 11:10:07.731071 47076539613632 model_training_utils.py:505] Train Step: 2045/2100 / loss = 0.82763671875 I0421 11:10:07.731515 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 4144 and 4145 I0421 11:10:08.812579 47076539613632 model_training_utils.py:505] Train Step: 2046/2100 / loss = 0.890625 I0421 11:10:08.813009 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.27 examples/second between steps 4145 and 4146 I0421 11:10:09.896678 47076539613632 model_training_utils.py:505] Train Step: 2047/2100 / loss = 1.13232421875 I0421 11:10:09.897109 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.97 examples/second between steps 4146 and 4147 I0421 11:10:10.979556 47076539613632 model_training_utils.py:505] Train Step: 2048/2100 / loss = 0.7572021484375 I0421 11:10:10.979988 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 4147 and 4148 I0421 11:10:12.062613 47076539613632 model_training_utils.py:505] Train Step: 2049/2100 / loss = 0.93896484375 I0421 11:10:12.063056 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.10 examples/second between steps 4148 and 4149 I0421 11:10:13.145261 47076539613632 model_training_utils.py:505] Train Step: 2050/2100 / loss = 1.20458984375 I0421 11:10:13.145696 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 4149 and 4150 I0421 11:10:14.228719 47076539613632 model_training_utils.py:505] Train Step: 2051/2100 / loss = 2.08154296875 I0421 11:10:14.229145 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 4150 and 4151 I0421 11:10:15.309481 47076539613632 model_training_utils.py:505] Train Step: 2052/2100 / loss = 1.566162109375 I0421 11:10:15.309906 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.34 examples/second between steps 4151 and 4152 I0421 11:10:16.393759 47076539613632 model_training_utils.py:505] Train Step: 2053/2100 / loss = 0.9586181640625 I0421 11:10:16.394189 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 4152 and 4153 I0421 11:10:17.476194 47076539613632 model_training_utils.py:505] Train Step: 2054/2100 / loss = 1.1304931640625 I0421 11:10:17.476626 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 4153 and 4154 I0421 11:10:18.560918 47076539613632 model_training_utils.py:505] Train Step: 2055/2100 / loss = 1.16796875 I0421 11:10:18.561352 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.92 examples/second between steps 4154 and 4155 I0421 11:10:19.642946 47076539613632 model_training_utils.py:505] Train Step: 2056/2100 / loss = 1.1414794921875 I0421 11:10:19.643382 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.21 examples/second between steps 4155 and 4156 I0421 11:10:20.726971 47076539613632 model_training_utils.py:505] Train Step: 2057/2100 / loss = 1.0753173828125 I0421 11:10:20.727403 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.99 examples/second between steps 4156 and 4157 I0421 11:10:21.809421 47076539613632 model_training_utils.py:505] Train Step: 2058/2100 / loss = 0.8421630859375 I0421 11:10:21.809843 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 4157 and 4158 I0421 11:10:22.894057 47076539613632 model_training_utils.py:505] Train Step: 2059/2100 / loss = 0.7177734375 I0421 11:10:22.894489 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.91 examples/second between steps 4158 and 4159 I0421 11:10:23.979171 47076539613632 model_training_utils.py:505] Train Step: 2060/2100 / loss = 0.975341796875 I0421 11:10:23.979603 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.89 examples/second between steps 4159 and 4160 I0421 11:10:25.060778 47076539613632 model_training_utils.py:505] Train Step: 2061/2100 / loss = 1.1600341796875 I0421 11:10:25.061204 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.27 examples/second between steps 4160 and 4161 I0421 11:10:26.143264 47076539613632 model_training_utils.py:505] Train Step: 2062/2100 / loss = 1.182373046875 I0421 11:10:26.143698 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.16 examples/second between steps 4161 and 4162 I0421 11:10:27.228803 47076539613632 model_training_utils.py:505] Train Step: 2063/2100 / loss = 0.764892578125 I0421 11:10:27.229224 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.83 examples/second between steps 4162 and 4163 I0421 11:10:28.313107 47076539613632 model_training_utils.py:505] Train Step: 2064/2100 / loss = 0.914306640625 I0421 11:10:28.313544 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 4163 and 4164 I0421 11:10:29.397438 47076539613632 model_training_utils.py:505] Train Step: 2065/2100 / loss = 0.850830078125 I0421 11:10:29.397867 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.95 examples/second between steps 4164 and 4165 I0421 11:10:30.485308 47076539613632 model_training_utils.py:505] Train Step: 2066/2100 / loss = 0.807861328125 I0421 11:10:30.485779 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.59 examples/second between steps 4165 and 4166 I0421 11:10:31.570574 47076539613632 model_training_utils.py:505] Train Step: 2067/2100 / loss = 0.9683837890625 I0421 11:10:31.571008 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 4166 and 4167 I0421 11:10:32.658983 47076539613632 model_training_utils.py:505] Train Step: 2068/2100 / loss = 0.8865966796875 I0421 11:10:32.659417 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.51 examples/second between steps 4167 and 4168 I0421 11:10:33.745277 47076539613632 model_training_utils.py:505] Train Step: 2069/2100 / loss = 0.9188232421875 I0421 11:10:33.745709 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.73 examples/second between steps 4168 and 4169 I0421 11:10:34.830404 47076539613632 model_training_utils.py:505] Train Step: 2070/2100 / loss = 1.334716796875 I0421 11:10:34.830834 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 4169 and 4170 I0421 11:10:35.914053 47076539613632 model_training_utils.py:505] Train Step: 2071/2100 / loss = 1.119873046875 I0421 11:10:35.914516 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 4170 and 4171 I0421 11:10:36.996649 47076539613632 model_training_utils.py:505] Train Step: 2072/2100 / loss = 1.35888671875 I0421 11:10:36.997076 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 4171 and 4172 I0421 11:10:38.074996 47076539613632 model_training_utils.py:505] Train Step: 2073/2100 / loss = 1.26708984375 I0421 11:10:38.075427 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.60 examples/second between steps 4172 and 4173 I0421 11:10:39.157679 47076539613632 model_training_utils.py:505] Train Step: 2074/2100 / loss = 1.0216064453125 I0421 11:10:39.158097 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.17 examples/second between steps 4173 and 4174 I0421 11:10:40.241211 47076539613632 model_training_utils.py:505] Train Step: 2075/2100 / loss = 0.8272705078125 I0421 11:10:40.241637 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 4174 and 4175 I0421 11:10:41.321363 47076539613632 model_training_utils.py:505] Train Step: 2076/2100 / loss = 0.7752685546875 I0421 11:10:41.321793 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.40 examples/second between steps 4175 and 4176 I0421 11:10:42.398024 47076539613632 model_training_utils.py:505] Train Step: 2077/2100 / loss = 0.7203369140625 I0421 11:10:42.398473 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.80 examples/second between steps 4176 and 4177 I0421 11:10:43.479971 47076539613632 model_training_utils.py:505] Train Step: 2078/2100 / loss = 0.7120361328125 I0421 11:10:43.480413 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 4177 and 4178 I0421 11:10:44.562592 47076539613632 model_training_utils.py:505] Train Step: 2079/2100 / loss = 0.70703125 I0421 11:10:44.563020 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.18 examples/second between steps 4178 and 4179 I0421 11:10:45.646338 47076539613632 model_training_utils.py:505] Train Step: 2080/2100 / loss = 0.821533203125 I0421 11:10:45.646767 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.02 examples/second between steps 4179 and 4180 I0421 11:10:46.727996 47076539613632 model_training_utils.py:505] Train Step: 2081/2100 / loss = 0.861328125 I0421 11:10:46.728390 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.28 examples/second between steps 4180 and 4181 I0421 11:10:47.810478 47076539613632 model_training_utils.py:505] Train Step: 2082/2100 / loss = 1.1680908203125 I0421 11:10:47.810855 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 4181 and 4182 I0421 11:10:48.892972 47076539613632 model_training_utils.py:505] Train Step: 2083/2100 / loss = 0.959716796875 I0421 11:10:48.893362 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.12 examples/second between steps 4182 and 4183 I0421 11:10:49.976273 47076539613632 model_training_utils.py:505] Train Step: 2084/2100 / loss = 0.7816162109375 I0421 11:10:49.976664 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 4183 and 4184 I0421 11:10:51.057787 47076539613632 model_training_utils.py:505] Train Step: 2085/2100 / loss = 0.766357421875 I0421 11:10:51.058168 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.25 examples/second between steps 4184 and 4185 I0421 11:10:52.141093 47076539613632 model_training_utils.py:505] Train Step: 2086/2100 / loss = 0.7110595703125 I0421 11:10:52.141477 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.05 examples/second between steps 4185 and 4186 I0421 11:10:53.223529 47076539613632 model_training_utils.py:505] Train Step: 2087/2100 / loss = 0.828857421875 I0421 11:10:53.223908 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.15 examples/second between steps 4186 and 4187 I0421 11:10:54.304207 47076539613632 model_training_utils.py:505] Train Step: 2088/2100 / loss = 0.8360595703125 I0421 11:10:54.304594 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.33 examples/second between steps 4187 and 4188 I0421 11:10:55.385206 47076539613632 model_training_utils.py:505] Train Step: 2089/2100 / loss = 0.92626953125 I0421 11:10:55.385594 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.32 examples/second between steps 4188 and 4189 I0421 11:10:56.470210 47076539613632 model_training_utils.py:505] Train Step: 2090/2100 / loss = 1.056884765625 I0421 11:10:56.470606 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.87 examples/second between steps 4189 and 4190 I0421 11:10:57.548849 47076539613632 model_training_utils.py:505] Train Step: 2091/2100 / loss = 1.41943359375 I0421 11:10:57.549230 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.57 examples/second between steps 4190 and 4191 I0421 11:10:58.629612 47076539613632 model_training_utils.py:505] Train Step: 2092/2100 / loss = 1.02587890625 I0421 11:10:58.629995 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.32 examples/second between steps 4191 and 4192 I0421 11:10:59.713944 47076539613632 model_training_utils.py:505] Train Step: 2093/2100 / loss = 1.027099609375 I0421 11:10:59.714338 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.94 examples/second between steps 4192 and 4193 I0421 11:11:00.795767 47076539613632 model_training_utils.py:505] Train Step: 2094/2100 / loss = 0.7979736328125 I0421 11:11:00.796154 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.24 examples/second between steps 4193 and 4194 I0421 11:11:01.874618 47076539613632 model_training_utils.py:505] Train Step: 2095/2100 / loss = 0.97607421875 I0421 11:11:01.875003 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.57 examples/second between steps 4194 and 4195 I0421 11:11:02.958481 47076539613632 model_training_utils.py:505] Train Step: 2096/2100 / loss = 0.8311767578125 I0421 11:11:02.958863 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.01 examples/second between steps 4195 and 4196 2021-04-21 11:11:04.027093: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. I0421 11:11:04.040890 47076539613632 model_training_utils.py:505] Train Step: 2097/2100 / loss = 0.5765380859375 I0421 11:11:04.041272 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.14 examples/second between steps 4196 and 4197 2021-04-21 11:11:04.142669: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. I0421 11:11:05.124349 47076539613632 model_training_utils.py:505] Train Step: 2098/2100 / loss = 0.8553466796875 I0421 11:11:05.124726 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 119.06 examples/second between steps 4197 and 4198 I0421 11:11:06.204141 47076539613632 model_training_utils.py:505] Train Step: 2099/2100 / loss = 0.859619140625 I0421 11:11:06.204528 47076539613632 keras_utils.py:133] TimeHistory: 1.07 seconds, 119.46 examples/second between steps 4198 and 4199 I0421 11:11:07.288696 47076539613632 model_training_utils.py:505] Train Step: 2100/2100 / loss = 0.77783203125 I0421 11:11:07.289077 47076539613632 keras_utils.py:133] TimeHistory: 1.08 seconds, 118.93 examples/second between steps 4199 and 4200 I0421 11:11:15.115145 47076539613632 model_training_utils.py:49] Saving model as TF checkpoint: /public/home/xuanbaby/DL-TensorFlow/models_r2.3.0/official/nlp/bert/model_squad_v2/ctl_step_2100.ckpt-3 I0421 11:11:15.143852 47076539613632 model_training_utils.py:97] Training Summary: {'total_training_steps': 2100, 'train_loss': 0.77783203125} I0421 11:11:20.734385 47076539613632 run_squad_helper.py:179] Restoring checkpoints from /public/home/xuanbaby/DL-TensorFlow/models_r2.3.0/official/nlp/bert/model_squad_v2/ctl_step_2100.ckpt-3 I0421 11:11:29.110588 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.110844 47076539613632 squad_lib.py:367] unique_id: 1000000000 I0421 11:11:29.110971 47076539613632 squad_lib.py:368] example_index: 0 I0421 11:11:29.111083 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.111301 47076539613632 squad_lib.py:371] tokens: [CLS] which n ##f ##l team represented the a ##f ##c at super bowl 50 ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.111489 47076539613632 squad_lib.py:374] token_to_orig_map: 17:0 18:1 19:2 20:3 21:4 22:5 23:5 24:5 25:6 26:7 27:8 28:9 29:10 30:11 31:12 32:13 33:14 34:15 35:16 36:17 37:17 38:17 39:17 40:17 41:18 42:19 43:20 44:21 45:21 46:22 47:23 48:23 49:23 50:24 51:25 52:26 53:26 54:26 55:26 56:26 57:27 58:28 59:28 60:29 61:29 62:29 63:30 64:31 65:32 66:33 67:34 68:35 69:35 70:35 71:35 72:35 73:36 74:37 75:37 76:37 77:38 78:38 79:38 80:39 81:39 82:39 83:40 84:41 85:42 86:43 87:44 88:45 89:46 90:46 91:47 92:48 93:49 94:50 95:51 96:52 97:52 98:52 99:52 100:53 101:53 102:54 103:54 104:55 105:56 106:56 107:56 108:56 109:57 110:58 111:59 112:60 113:60 114:61 115:61 116:61 117:61 118:62 119:63 120:64 121:65 122:65 123:66 124:66 125:66 126:66 127:67 128:67 129:67 130:67 131:67 132:67 133:68 134:69 135:70 136:71 137:72 138:73 139:74 140:74 141:75 142:76 143:77 144:78 145:79 146:79 147:80 148:80 149:81 150:82 151:83 152:83 153:83 154:84 155:84 156:85 157:86 158:87 159:88 160:89 161:89 162:89 163:90 164:91 165:92 166:93 167:94 168:95 169:96 170:97 171:98 172:99 173:99 174:100 175:100 176:100 177:101 178:101 179:102 180:103 181:104 182:105 183:106 184:107 185:108 186:109 187:110 188:110 189:111 190:112 191:112 192:112 193:112 194:113 195:114 196:115 197:116 198:117 199:118 200:119 201:120 202:121 203:121 204:121 205:122 206:122 207:122 208:123 209:123 I0421 11:11:29.111668 47076539613632 squad_lib.py:379] token_is_max_context: 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True 204:True 205:True 206:True 207:True 208:True 209:True I0421 11:11:29.111902 47076539613632 squad_lib.py:381] input_ids: 101 1134 183 2087 1233 1264 2533 1103 170 2087 1665 1120 7688 7329 1851 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.112115 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.112339 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.151947 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.152087 47076539613632 squad_lib.py:367] unique_id: 1000000001 I0421 11:11:29.152208 47076539613632 squad_lib.py:368] example_index: 1 I0421 11:11:29.152330 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.152540 47076539613632 squad_lib.py:371] tokens: [CLS] which n ##f ##l team represented the n ##f ##c at super bowl 50 ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.152730 47076539613632 squad_lib.py:374] token_to_orig_map: 17:0 18:1 19:2 20:3 21:4 22:5 23:5 24:5 25:6 26:7 27:8 28:9 29:10 30:11 31:12 32:13 33:14 34:15 35:16 36:17 37:17 38:17 39:17 40:17 41:18 42:19 43:20 44:21 45:21 46:22 47:23 48:23 49:23 50:24 51:25 52:26 53:26 54:26 55:26 56:26 57:27 58:28 59:28 60:29 61:29 62:29 63:30 64:31 65:32 66:33 67:34 68:35 69:35 70:35 71:35 72:35 73:36 74:37 75:37 76:37 77:38 78:38 79:38 80:39 81:39 82:39 83:40 84:41 85:42 86:43 87:44 88:45 89:46 90:46 91:47 92:48 93:49 94:50 95:51 96:52 97:52 98:52 99:52 100:53 101:53 102:54 103:54 104:55 105:56 106:56 107:56 108:56 109:57 110:58 111:59 112:60 113:60 114:61 115:61 116:61 117:61 118:62 119:63 120:64 121:65 122:65 123:66 124:66 125:66 126:66 127:67 128:67 129:67 130:67 131:67 132:67 133:68 134:69 135:70 136:71 137:72 138:73 139:74 140:74 141:75 142:76 143:77 144:78 145:79 146:79 147:80 148:80 149:81 150:82 151:83 152:83 153:83 154:84 155:84 156:85 157:86 158:87 159:88 160:89 161:89 162:89 163:90 164:91 165:92 166:93 167:94 168:95 169:96 170:97 171:98 172:99 173:99 174:100 175:100 176:100 177:101 178:101 179:102 180:103 181:104 182:105 183:106 184:107 185:108 186:109 187:110 188:110 189:111 190:112 191:112 192:112 193:112 194:113 195:114 196:115 197:116 198:117 199:118 200:119 201:120 202:121 203:121 204:121 205:122 206:122 207:122 208:123 209:123 I0421 11:11:29.152909 47076539613632 squad_lib.py:379] token_is_max_context: 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True 204:True 205:True 206:True 207:True 208:True 209:True I0421 11:11:29.153126 47076539613632 squad_lib.py:381] input_ids: 101 1134 183 2087 1233 1264 2533 1103 183 2087 1665 1120 7688 7329 1851 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.153357 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.153572 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.159307 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.159441 47076539613632 squad_lib.py:367] unique_id: 1000000002 I0421 11:11:29.159569 47076539613632 squad_lib.py:368] example_index: 2 I0421 11:11:29.159681 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.159886 47076539613632 squad_lib.py:371] tokens: [CLS] where did super bowl 50 take place ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.160074 47076539613632 squad_lib.py:374] token_to_orig_map: 10:0 11:1 12:2 13:3 14:4 15:5 16:5 17:5 18:6 19:7 20:8 21:9 22:10 23:11 24:12 25:13 26:14 27:15 28:16 29:17 30:17 31:17 32:17 33:17 34:18 35:19 36:20 37:21 38:21 39:22 40:23 41:23 42:23 43:24 44:25 45:26 46:26 47:26 48:26 49:26 50:27 51:28 52:28 53:29 54:29 55:29 56:30 57:31 58:32 59:33 60:34 61:35 62:35 63:35 64:35 65:35 66:36 67:37 68:37 69:37 70:38 71:38 72:38 73:39 74:39 75:39 76:40 77:41 78:42 79:43 80:44 81:45 82:46 83:46 84:47 85:48 86:49 87:50 88:51 89:52 90:52 91:52 92:52 93:53 94:53 95:54 96:54 97:55 98:56 99:56 100:56 101:56 102:57 103:58 104:59 105:60 106:60 107:61 108:61 109:61 110:61 111:62 112:63 113:64 114:65 115:65 116:66 117:66 118:66 119:66 120:67 121:67 122:67 123:67 124:67 125:67 126:68 127:69 128:70 129:71 130:72 131:73 132:74 133:74 134:75 135:76 136:77 137:78 138:79 139:79 140:80 141:80 142:81 143:82 144:83 145:83 146:83 147:84 148:84 149:85 150:86 151:87 152:88 153:89 154:89 155:89 156:90 157:91 158:92 159:93 160:94 161:95 162:96 163:97 164:98 165:99 166:99 167:100 168:100 169:100 170:101 171:101 172:102 173:103 174:104 175:105 176:106 177:107 178:108 179:109 180:110 181:110 182:111 183:112 184:112 185:112 186:112 187:113 188:114 189:115 190:116 191:117 192:118 193:119 194:120 195:121 196:121 197:121 198:122 199:122 200:122 201:123 202:123 I0421 11:11:29.160250 47076539613632 squad_lib.py:379] token_is_max_context: 10:True 11:True 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True I0421 11:11:29.160475 47076539613632 squad_lib.py:381] input_ids: 101 1187 1225 7688 7329 1851 1321 1282 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.160690 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.160901 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.166612 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.166741 47076539613632 squad_lib.py:367] unique_id: 1000000003 I0421 11:11:29.166857 47076539613632 squad_lib.py:368] example_index: 3 I0421 11:11:29.166969 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.167174 47076539613632 squad_lib.py:371] tokens: [CLS] which n ##f ##l team won super bowl 50 ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.167364 47076539613632 squad_lib.py:374] token_to_orig_map: 12:0 13:1 14:2 15:3 16:4 17:5 18:5 19:5 20:6 21:7 22:8 23:9 24:10 25:11 26:12 27:13 28:14 29:15 30:16 31:17 32:17 33:17 34:17 35:17 36:18 37:19 38:20 39:21 40:21 41:22 42:23 43:23 44:23 45:24 46:25 47:26 48:26 49:26 50:26 51:26 52:27 53:28 54:28 55:29 56:29 57:29 58:30 59:31 60:32 61:33 62:34 63:35 64:35 65:35 66:35 67:35 68:36 69:37 70:37 71:37 72:38 73:38 74:38 75:39 76:39 77:39 78:40 79:41 80:42 81:43 82:44 83:45 84:46 85:46 86:47 87:48 88:49 89:50 90:51 91:52 92:52 93:52 94:52 95:53 96:53 97:54 98:54 99:55 100:56 101:56 102:56 103:56 104:57 105:58 106:59 107:60 108:60 109:61 110:61 111:61 112:61 113:62 114:63 115:64 116:65 117:65 118:66 119:66 120:66 121:66 122:67 123:67 124:67 125:67 126:67 127:67 128:68 129:69 130:70 131:71 132:72 133:73 134:74 135:74 136:75 137:76 138:77 139:78 140:79 141:79 142:80 143:80 144:81 145:82 146:83 147:83 148:83 149:84 150:84 151:85 152:86 153:87 154:88 155:89 156:89 157:89 158:90 159:91 160:92 161:93 162:94 163:95 164:96 165:97 166:98 167:99 168:99 169:100 170:100 171:100 172:101 173:101 174:102 175:103 176:104 177:105 178:106 179:107 180:108 181:109 182:110 183:110 184:111 185:112 186:112 187:112 188:112 189:113 190:114 191:115 192:116 193:117 194:118 195:119 196:120 197:121 198:121 199:121 200:122 201:122 202:122 203:123 204:123 I0421 11:11:29.167560 47076539613632 squad_lib.py:379] token_is_max_context: 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True 204:True I0421 11:11:29.167780 47076539613632 squad_lib.py:381] input_ids: 101 1134 183 2087 1233 1264 1281 7688 7329 1851 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.167994 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.168213 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.174062 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.174190 47076539613632 squad_lib.py:367] unique_id: 1000000004 I0421 11:11:29.174315 47076539613632 squad_lib.py:368] example_index: 4 I0421 11:11:29.174427 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.174633 47076539613632 squad_lib.py:371] tokens: [CLS] what color was used to emphasize the 50th anniversary of the super bowl ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.174815 47076539613632 squad_lib.py:374] token_to_orig_map: 16:0 17:1 18:2 19:3 20:4 21:5 22:5 23:5 24:6 25:7 26:8 27:9 28:10 29:11 30:12 31:13 32:14 33:15 34:16 35:17 36:17 37:17 38:17 39:17 40:18 41:19 42:20 43:21 44:21 45:22 46:23 47:23 48:23 49:24 50:25 51:26 52:26 53:26 54:26 55:26 56:27 57:28 58:28 59:29 60:29 61:29 62:30 63:31 64:32 65:33 66:34 67:35 68:35 69:35 70:35 71:35 72:36 73:37 74:37 75:37 76:38 77:38 78:38 79:39 80:39 81:39 82:40 83:41 84:42 85:43 86:44 87:45 88:46 89:46 90:47 91:48 92:49 93:50 94:51 95:52 96:52 97:52 98:52 99:53 100:53 101:54 102:54 103:55 104:56 105:56 106:56 107:56 108:57 109:58 110:59 111:60 112:60 113:61 114:61 115:61 116:61 117:62 118:63 119:64 120:65 121:65 122:66 123:66 124:66 125:66 126:67 127:67 128:67 129:67 130:67 131:67 132:68 133:69 134:70 135:71 136:72 137:73 138:74 139:74 140:75 141:76 142:77 143:78 144:79 145:79 146:80 147:80 148:81 149:82 150:83 151:83 152:83 153:84 154:84 155:85 156:86 157:87 158:88 159:89 160:89 161:89 162:90 163:91 164:92 165:93 166:94 167:95 168:96 169:97 170:98 171:99 172:99 173:100 174:100 175:100 176:101 177:101 178:102 179:103 180:104 181:105 182:106 183:107 184:108 185:109 186:110 187:110 188:111 189:112 190:112 191:112 192:112 193:113 194:114 195:115 196:116 197:117 198:118 199:119 200:120 201:121 202:121 203:121 204:122 205:122 206:122 207:123 208:123 I0421 11:11:29.175000 47076539613632 squad_lib.py:379] token_is_max_context: 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True 204:True 205:True 206:True 207:True 208:True I0421 11:11:29.175219 47076539613632 squad_lib.py:381] input_ids: 101 1184 2942 1108 1215 1106 19291 1103 13163 5453 1104 1103 7688 7329 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.175444 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.175664 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.181401 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.181531 47076539613632 squad_lib.py:367] unique_id: 1000000005 I0421 11:11:29.181646 47076539613632 squad_lib.py:368] example_index: 5 I0421 11:11:29.181756 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.181961 47076539613632 squad_lib.py:371] tokens: [CLS] what was the theme of super bowl 50 ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.182143 47076539613632 squad_lib.py:374] token_to_orig_map: 11:0 12:1 13:2 14:3 15:4 16:5 17:5 18:5 19:6 20:7 21:8 22:9 23:10 24:11 25:12 26:13 27:14 28:15 29:16 30:17 31:17 32:17 33:17 34:17 35:18 36:19 37:20 38:21 39:21 40:22 41:23 42:23 43:23 44:24 45:25 46:26 47:26 48:26 49:26 50:26 51:27 52:28 53:28 54:29 55:29 56:29 57:30 58:31 59:32 60:33 61:34 62:35 63:35 64:35 65:35 66:35 67:36 68:37 69:37 70:37 71:38 72:38 73:38 74:39 75:39 76:39 77:40 78:41 79:42 80:43 81:44 82:45 83:46 84:46 85:47 86:48 87:49 88:50 89:51 90:52 91:52 92:52 93:52 94:53 95:53 96:54 97:54 98:55 99:56 100:56 101:56 102:56 103:57 104:58 105:59 106:60 107:60 108:61 109:61 110:61 111:61 112:62 113:63 114:64 115:65 116:65 117:66 118:66 119:66 120:66 121:67 122:67 123:67 124:67 125:67 126:67 127:68 128:69 129:70 130:71 131:72 132:73 133:74 134:74 135:75 136:76 137:77 138:78 139:79 140:79 141:80 142:80 143:81 144:82 145:83 146:83 147:83 148:84 149:84 150:85 151:86 152:87 153:88 154:89 155:89 156:89 157:90 158:91 159:92 160:93 161:94 162:95 163:96 164:97 165:98 166:99 167:99 168:100 169:100 170:100 171:101 172:101 173:102 174:103 175:104 176:105 177:106 178:107 179:108 180:109 181:110 182:110 183:111 184:112 185:112 186:112 187:112 188:113 189:114 190:115 191:116 192:117 193:118 194:119 195:120 196:121 197:121 198:121 199:122 200:122 201:122 202:123 203:123 I0421 11:11:29.182329 47076539613632 squad_lib.py:379] token_is_max_context: 11:True 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True I0421 11:11:29.182553 47076539613632 squad_lib.py:381] input_ids: 101 1184 1108 1103 3815 1104 7688 7329 1851 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.182765 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.182975 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.188714 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.188843 47076539613632 squad_lib.py:367] unique_id: 1000000006 I0421 11:11:29.188958 47076539613632 squad_lib.py:368] example_index: 6 I0421 11:11:29.189069 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.189288 47076539613632 squad_lib.py:371] tokens: [CLS] what day was the game played on ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.189475 47076539613632 squad_lib.py:374] token_to_orig_map: 10:0 11:1 12:2 13:3 14:4 15:5 16:5 17:5 18:6 19:7 20:8 21:9 22:10 23:11 24:12 25:13 26:14 27:15 28:16 29:17 30:17 31:17 32:17 33:17 34:18 35:19 36:20 37:21 38:21 39:22 40:23 41:23 42:23 43:24 44:25 45:26 46:26 47:26 48:26 49:26 50:27 51:28 52:28 53:29 54:29 55:29 56:30 57:31 58:32 59:33 60:34 61:35 62:35 63:35 64:35 65:35 66:36 67:37 68:37 69:37 70:38 71:38 72:38 73:39 74:39 75:39 76:40 77:41 78:42 79:43 80:44 81:45 82:46 83:46 84:47 85:48 86:49 87:50 88:51 89:52 90:52 91:52 92:52 93:53 94:53 95:54 96:54 97:55 98:56 99:56 100:56 101:56 102:57 103:58 104:59 105:60 106:60 107:61 108:61 109:61 110:61 111:62 112:63 113:64 114:65 115:65 116:66 117:66 118:66 119:66 120:67 121:67 122:67 123:67 124:67 125:67 126:68 127:69 128:70 129:71 130:72 131:73 132:74 133:74 134:75 135:76 136:77 137:78 138:79 139:79 140:80 141:80 142:81 143:82 144:83 145:83 146:83 147:84 148:84 149:85 150:86 151:87 152:88 153:89 154:89 155:89 156:90 157:91 158:92 159:93 160:94 161:95 162:96 163:97 164:98 165:99 166:99 167:100 168:100 169:100 170:101 171:101 172:102 173:103 174:104 175:105 176:106 177:107 178:108 179:109 180:110 181:110 182:111 183:112 184:112 185:112 186:112 187:113 188:114 189:115 190:116 191:117 192:118 193:119 194:120 195:121 196:121 197:121 198:122 199:122 200:122 201:123 202:123 I0421 11:11:29.189651 47076539613632 squad_lib.py:379] token_is_max_context: 10:True 11:True 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True I0421 11:11:29.189876 47076539613632 squad_lib.py:381] input_ids: 101 1184 1285 1108 1103 1342 1307 1113 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.190092 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.190315 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.195949 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.196079 47076539613632 squad_lib.py:367] unique_id: 1000000007 I0421 11:11:29.196193 47076539613632 squad_lib.py:368] example_index: 7 I0421 11:11:29.196310 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.196518 47076539613632 squad_lib.py:371] tokens: [CLS] what is the a ##f ##c short for ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.196707 47076539613632 squad_lib.py:374] token_to_orig_map: 11:0 12:1 13:2 14:3 15:4 16:5 17:5 18:5 19:6 20:7 21:8 22:9 23:10 24:11 25:12 26:13 27:14 28:15 29:16 30:17 31:17 32:17 33:17 34:17 35:18 36:19 37:20 38:21 39:21 40:22 41:23 42:23 43:23 44:24 45:25 46:26 47:26 48:26 49:26 50:26 51:27 52:28 53:28 54:29 55:29 56:29 57:30 58:31 59:32 60:33 61:34 62:35 63:35 64:35 65:35 66:35 67:36 68:37 69:37 70:37 71:38 72:38 73:38 74:39 75:39 76:39 77:40 78:41 79:42 80:43 81:44 82:45 83:46 84:46 85:47 86:48 87:49 88:50 89:51 90:52 91:52 92:52 93:52 94:53 95:53 96:54 97:54 98:55 99:56 100:56 101:56 102:56 103:57 104:58 105:59 106:60 107:60 108:61 109:61 110:61 111:61 112:62 113:63 114:64 115:65 116:65 117:66 118:66 119:66 120:66 121:67 122:67 123:67 124:67 125:67 126:67 127:68 128:69 129:70 130:71 131:72 132:73 133:74 134:74 135:75 136:76 137:77 138:78 139:79 140:79 141:80 142:80 143:81 144:82 145:83 146:83 147:83 148:84 149:84 150:85 151:86 152:87 153:88 154:89 155:89 156:89 157:90 158:91 159:92 160:93 161:94 162:95 163:96 164:97 165:98 166:99 167:99 168:100 169:100 170:100 171:101 172:101 173:102 174:103 175:104 176:105 177:106 178:107 179:108 180:109 181:110 182:110 183:111 184:112 185:112 186:112 187:112 188:113 189:114 190:115 191:116 192:117 193:118 194:119 195:120 196:121 197:121 198:121 199:122 200:122 201:122 202:123 203:123 I0421 11:11:29.196887 47076539613632 squad_lib.py:379] token_is_max_context: 11:True 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True I0421 11:11:29.197104 47076539613632 squad_lib.py:381] input_ids: 101 1184 1110 1103 170 2087 1665 1603 1111 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.197331 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.197552 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.203239 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.203375 47076539613632 squad_lib.py:367] unique_id: 1000000008 I0421 11:11:29.203488 47076539613632 squad_lib.py:368] example_index: 8 I0421 11:11:29.203599 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.203802 47076539613632 squad_lib.py:371] tokens: [CLS] what was the theme of super bowl 50 ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.203990 47076539613632 squad_lib.py:374] token_to_orig_map: 11:0 12:1 13:2 14:3 15:4 16:5 17:5 18:5 19:6 20:7 21:8 22:9 23:10 24:11 25:12 26:13 27:14 28:15 29:16 30:17 31:17 32:17 33:17 34:17 35:18 36:19 37:20 38:21 39:21 40:22 41:23 42:23 43:23 44:24 45:25 46:26 47:26 48:26 49:26 50:26 51:27 52:28 53:28 54:29 55:29 56:29 57:30 58:31 59:32 60:33 61:34 62:35 63:35 64:35 65:35 66:35 67:36 68:37 69:37 70:37 71:38 72:38 73:38 74:39 75:39 76:39 77:40 78:41 79:42 80:43 81:44 82:45 83:46 84:46 85:47 86:48 87:49 88:50 89:51 90:52 91:52 92:52 93:52 94:53 95:53 96:54 97:54 98:55 99:56 100:56 101:56 102:56 103:57 104:58 105:59 106:60 107:60 108:61 109:61 110:61 111:61 112:62 113:63 114:64 115:65 116:65 117:66 118:66 119:66 120:66 121:67 122:67 123:67 124:67 125:67 126:67 127:68 128:69 129:70 130:71 131:72 132:73 133:74 134:74 135:75 136:76 137:77 138:78 139:79 140:79 141:80 142:80 143:81 144:82 145:83 146:83 147:83 148:84 149:84 150:85 151:86 152:87 153:88 154:89 155:89 156:89 157:90 158:91 159:92 160:93 161:94 162:95 163:96 164:97 165:98 166:99 167:99 168:100 169:100 170:100 171:101 172:101 173:102 174:103 175:104 176:105 177:106 178:107 179:108 180:109 181:110 182:110 183:111 184:112 185:112 186:112 187:112 188:113 189:114 190:115 191:116 192:117 193:118 194:119 195:120 196:121 197:121 198:121 199:122 200:122 201:122 202:123 203:123 I0421 11:11:29.204166 47076539613632 squad_lib.py:379] token_is_max_context: 11:True 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True I0421 11:11:29.204387 47076539613632 squad_lib.py:381] input_ids: 101 1184 1108 1103 3815 1104 7688 7329 1851 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.204607 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.204820 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.210498 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.210627 47076539613632 squad_lib.py:367] unique_id: 1000000009 I0421 11:11:29.210741 47076539613632 squad_lib.py:368] example_index: 9 I0421 11:11:29.210862 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.211064 47076539613632 squad_lib.py:371] tokens: [CLS] what does a ##f ##c stand for ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.211248 47076539613632 squad_lib.py:374] token_to_orig_map: 10:0 11:1 12:2 13:3 14:4 15:5 16:5 17:5 18:6 19:7 20:8 21:9 22:10 23:11 24:12 25:13 26:14 27:15 28:16 29:17 30:17 31:17 32:17 33:17 34:18 35:19 36:20 37:21 38:21 39:22 40:23 41:23 42:23 43:24 44:25 45:26 46:26 47:26 48:26 49:26 50:27 51:28 52:28 53:29 54:29 55:29 56:30 57:31 58:32 59:33 60:34 61:35 62:35 63:35 64:35 65:35 66:36 67:37 68:37 69:37 70:38 71:38 72:38 73:39 74:39 75:39 76:40 77:41 78:42 79:43 80:44 81:45 82:46 83:46 84:47 85:48 86:49 87:50 88:51 89:52 90:52 91:52 92:52 93:53 94:53 95:54 96:54 97:55 98:56 99:56 100:56 101:56 102:57 103:58 104:59 105:60 106:60 107:61 108:61 109:61 110:61 111:62 112:63 113:64 114:65 115:65 116:66 117:66 118:66 119:66 120:67 121:67 122:67 123:67 124:67 125:67 126:68 127:69 128:70 129:71 130:72 131:73 132:74 133:74 134:75 135:76 136:77 137:78 138:79 139:79 140:80 141:80 142:81 143:82 144:83 145:83 146:83 147:84 148:84 149:85 150:86 151:87 152:88 153:89 154:89 155:89 156:90 157:91 158:92 159:93 160:94 161:95 162:96 163:97 164:98 165:99 166:99 167:100 168:100 169:100 170:101 171:101 172:102 173:103 174:104 175:105 176:106 177:107 178:108 179:109 180:110 181:110 182:111 183:112 184:112 185:112 186:112 187:113 188:114 189:115 190:116 191:117 192:118 193:119 194:120 195:121 196:121 197:121 198:122 199:122 200:122 201:123 202:123 I0421 11:11:29.211440 47076539613632 squad_lib.py:379] token_is_max_context: 10:True 11:True 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True I0421 11:11:29.211653 47076539613632 squad_lib.py:381] input_ids: 101 1184 1674 170 2087 1665 2484 1111 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.211866 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.212077 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.217821 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.217953 47076539613632 squad_lib.py:367] unique_id: 1000000010 I0421 11:11:29.218068 47076539613632 squad_lib.py:368] example_index: 10 I0421 11:11:29.218178 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.218395 47076539613632 squad_lib.py:371] tokens: [CLS] what day was the super bowl played on ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.218580 47076539613632 squad_lib.py:374] token_to_orig_map: 11:0 12:1 13:2 14:3 15:4 16:5 17:5 18:5 19:6 20:7 21:8 22:9 23:10 24:11 25:12 26:13 27:14 28:15 29:16 30:17 31:17 32:17 33:17 34:17 35:18 36:19 37:20 38:21 39:21 40:22 41:23 42:23 43:23 44:24 45:25 46:26 47:26 48:26 49:26 50:26 51:27 52:28 53:28 54:29 55:29 56:29 57:30 58:31 59:32 60:33 61:34 62:35 63:35 64:35 65:35 66:35 67:36 68:37 69:37 70:37 71:38 72:38 73:38 74:39 75:39 76:39 77:40 78:41 79:42 80:43 81:44 82:45 83:46 84:46 85:47 86:48 87:49 88:50 89:51 90:52 91:52 92:52 93:52 94:53 95:53 96:54 97:54 98:55 99:56 100:56 101:56 102:56 103:57 104:58 105:59 106:60 107:60 108:61 109:61 110:61 111:61 112:62 113:63 114:64 115:65 116:65 117:66 118:66 119:66 120:66 121:67 122:67 123:67 124:67 125:67 126:67 127:68 128:69 129:70 130:71 131:72 132:73 133:74 134:74 135:75 136:76 137:77 138:78 139:79 140:79 141:80 142:80 143:81 144:82 145:83 146:83 147:83 148:84 149:84 150:85 151:86 152:87 153:88 154:89 155:89 156:89 157:90 158:91 159:92 160:93 161:94 162:95 163:96 164:97 165:98 166:99 167:99 168:100 169:100 170:100 171:101 172:101 173:102 174:103 175:104 176:105 177:106 178:107 179:108 180:109 181:110 182:110 183:111 184:112 185:112 186:112 187:112 188:113 189:114 190:115 191:116 192:117 193:118 194:119 195:120 196:121 197:121 198:121 199:122 200:122 201:122 202:123 203:123 I0421 11:11:29.218762 47076539613632 squad_lib.py:379] token_is_max_context: 11:True 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True I0421 11:11:29.218987 47076539613632 squad_lib.py:381] input_ids: 101 1184 1285 1108 1103 7688 7329 1307 1113 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.219200 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.219423 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.225099 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.225227 47076539613632 squad_lib.py:367] unique_id: 1000000011 I0421 11:11:29.225351 47076539613632 squad_lib.py:368] example_index: 11 I0421 11:11:29.225461 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.225666 47076539613632 squad_lib.py:371] tokens: [CLS] who won super bowl 50 ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.225848 47076539613632 squad_lib.py:374] token_to_orig_map: 8:0 9:1 10:2 11:3 12:4 13:5 14:5 15:5 16:6 17:7 18:8 19:9 20:10 21:11 22:12 23:13 24:14 25:15 26:16 27:17 28:17 29:17 30:17 31:17 32:18 33:19 34:20 35:21 36:21 37:22 38:23 39:23 40:23 41:24 42:25 43:26 44:26 45:26 46:26 47:26 48:27 49:28 50:28 51:29 52:29 53:29 54:30 55:31 56:32 57:33 58:34 59:35 60:35 61:35 62:35 63:35 64:36 65:37 66:37 67:37 68:38 69:38 70:38 71:39 72:39 73:39 74:40 75:41 76:42 77:43 78:44 79:45 80:46 81:46 82:47 83:48 84:49 85:50 86:51 87:52 88:52 89:52 90:52 91:53 92:53 93:54 94:54 95:55 96:56 97:56 98:56 99:56 100:57 101:58 102:59 103:60 104:60 105:61 106:61 107:61 108:61 109:62 110:63 111:64 112:65 113:65 114:66 115:66 116:66 117:66 118:67 119:67 120:67 121:67 122:67 123:67 124:68 125:69 126:70 127:71 128:72 129:73 130:74 131:74 132:75 133:76 134:77 135:78 136:79 137:79 138:80 139:80 140:81 141:82 142:83 143:83 144:83 145:84 146:84 147:85 148:86 149:87 150:88 151:89 152:89 153:89 154:90 155:91 156:92 157:93 158:94 159:95 160:96 161:97 162:98 163:99 164:99 165:100 166:100 167:100 168:101 169:101 170:102 171:103 172:104 173:105 174:106 175:107 176:108 177:109 178:110 179:110 180:111 181:112 182:112 183:112 184:112 185:113 186:114 187:115 188:116 189:117 190:118 191:119 192:120 193:121 194:121 195:121 196:122 197:122 198:122 199:123 200:123 I0421 11:11:29.226024 47076539613632 squad_lib.py:379] token_is_max_context: 8:True 9:True 10:True 11:True 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True I0421 11:11:29.226248 47076539613632 squad_lib.py:381] input_ids: 101 1150 1281 7688 7329 1851 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.226472 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.226676 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.232438 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.232580 47076539613632 squad_lib.py:367] unique_id: 1000000012 I0421 11:11:29.232693 47076539613632 squad_lib.py:368] example_index: 12 I0421 11:11:29.232803 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.233011 47076539613632 squad_lib.py:371] tokens: [CLS] what venue did super bowl 50 take place in ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.233192 47076539613632 squad_lib.py:374] token_to_orig_map: 12:0 13:1 14:2 15:3 16:4 17:5 18:5 19:5 20:6 21:7 22:8 23:9 24:10 25:11 26:12 27:13 28:14 29:15 30:16 31:17 32:17 33:17 34:17 35:17 36:18 37:19 38:20 39:21 40:21 41:22 42:23 43:23 44:23 45:24 46:25 47:26 48:26 49:26 50:26 51:26 52:27 53:28 54:28 55:29 56:29 57:29 58:30 59:31 60:32 61:33 62:34 63:35 64:35 65:35 66:35 67:35 68:36 69:37 70:37 71:37 72:38 73:38 74:38 75:39 76:39 77:39 78:40 79:41 80:42 81:43 82:44 83:45 84:46 85:46 86:47 87:48 88:49 89:50 90:51 91:52 92:52 93:52 94:52 95:53 96:53 97:54 98:54 99:55 100:56 101:56 102:56 103:56 104:57 105:58 106:59 107:60 108:60 109:61 110:61 111:61 112:61 113:62 114:63 115:64 116:65 117:65 118:66 119:66 120:66 121:66 122:67 123:67 124:67 125:67 126:67 127:67 128:68 129:69 130:70 131:71 132:72 133:73 134:74 135:74 136:75 137:76 138:77 139:78 140:79 141:79 142:80 143:80 144:81 145:82 146:83 147:83 148:83 149:84 150:84 151:85 152:86 153:87 154:88 155:89 156:89 157:89 158:90 159:91 160:92 161:93 162:94 163:95 164:96 165:97 166:98 167:99 168:99 169:100 170:100 171:100 172:101 173:101 174:102 175:103 176:104 177:105 178:106 179:107 180:108 181:109 182:110 183:110 184:111 185:112 186:112 187:112 188:112 189:113 190:114 191:115 192:116 193:117 194:118 195:119 196:120 197:121 198:121 199:121 200:122 201:122 202:122 203:123 204:123 I0421 11:11:29.233377 47076539613632 squad_lib.py:379] token_is_max_context: 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True 204:True I0421 11:11:29.233603 47076539613632 squad_lib.py:381] input_ids: 101 1184 6590 1225 7688 7329 1851 1321 1282 1107 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.233817 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.234028 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.239761 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.239892 47076539613632 squad_lib.py:367] unique_id: 1000000013 I0421 11:11:29.240006 47076539613632 squad_lib.py:368] example_index: 13 I0421 11:11:29.240114 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.240329 47076539613632 squad_lib.py:371] tokens: [CLS] what city did super bowl 50 take place in ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.240523 47076539613632 squad_lib.py:374] token_to_orig_map: 12:0 13:1 14:2 15:3 16:4 17:5 18:5 19:5 20:6 21:7 22:8 23:9 24:10 25:11 26:12 27:13 28:14 29:15 30:16 31:17 32:17 33:17 34:17 35:17 36:18 37:19 38:20 39:21 40:21 41:22 42:23 43:23 44:23 45:24 46:25 47:26 48:26 49:26 50:26 51:26 52:27 53:28 54:28 55:29 56:29 57:29 58:30 59:31 60:32 61:33 62:34 63:35 64:35 65:35 66:35 67:35 68:36 69:37 70:37 71:37 72:38 73:38 74:38 75:39 76:39 77:39 78:40 79:41 80:42 81:43 82:44 83:45 84:46 85:46 86:47 87:48 88:49 89:50 90:51 91:52 92:52 93:52 94:52 95:53 96:53 97:54 98:54 99:55 100:56 101:56 102:56 103:56 104:57 105:58 106:59 107:60 108:60 109:61 110:61 111:61 112:61 113:62 114:63 115:64 116:65 117:65 118:66 119:66 120:66 121:66 122:67 123:67 124:67 125:67 126:67 127:67 128:68 129:69 130:70 131:71 132:72 133:73 134:74 135:74 136:75 137:76 138:77 139:78 140:79 141:79 142:80 143:80 144:81 145:82 146:83 147:83 148:83 149:84 150:84 151:85 152:86 153:87 154:88 155:89 156:89 157:89 158:90 159:91 160:92 161:93 162:94 163:95 164:96 165:97 166:98 167:99 168:99 169:100 170:100 171:100 172:101 173:101 174:102 175:103 176:104 177:105 178:106 179:107 180:108 181:109 182:110 183:110 184:111 185:112 186:112 187:112 188:112 189:113 190:114 191:115 192:116 193:117 194:118 195:119 196:120 197:121 198:121 199:121 200:122 201:122 202:122 203:123 204:123 I0421 11:11:29.240707 47076539613632 squad_lib.py:379] token_is_max_context: 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True 204:True I0421 11:11:29.240931 47076539613632 squad_lib.py:381] input_ids: 101 1184 1331 1225 7688 7329 1851 1321 1282 1107 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.241142 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.241360 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.247224 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.247361 47076539613632 squad_lib.py:367] unique_id: 1000000014 I0421 11:11:29.247475 47076539613632 squad_lib.py:368] example_index: 14 I0421 11:11:29.247584 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.247793 47076539613632 squad_lib.py:371] tokens: [CLS] if r ##oman n ##ume ##rals were used , what would super bowl 50 have been called ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.247981 47076539613632 squad_lib.py:374] token_to_orig_map: 20:0 21:1 22:2 23:3 24:4 25:5 26:5 27:5 28:6 29:7 30:8 31:9 32:10 33:11 34:12 35:13 36:14 37:15 38:16 39:17 40:17 41:17 42:17 43:17 44:18 45:19 46:20 47:21 48:21 49:22 50:23 51:23 52:23 53:24 54:25 55:26 56:26 57:26 58:26 59:26 60:27 61:28 62:28 63:29 64:29 65:29 66:30 67:31 68:32 69:33 70:34 71:35 72:35 73:35 74:35 75:35 76:36 77:37 78:37 79:37 80:38 81:38 82:38 83:39 84:39 85:39 86:40 87:41 88:42 89:43 90:44 91:45 92:46 93:46 94:47 95:48 96:49 97:50 98:51 99:52 100:52 101:52 102:52 103:53 104:53 105:54 106:54 107:55 108:56 109:56 110:56 111:56 112:57 113:58 114:59 115:60 116:60 117:61 118:61 119:61 120:61 121:62 122:63 123:64 124:65 125:65 126:66 127:66 128:66 129:66 130:67 131:67 132:67 133:67 134:67 135:67 136:68 137:69 138:70 139:71 140:72 141:73 142:74 143:74 144:75 145:76 146:77 147:78 148:79 149:79 150:80 151:80 152:81 153:82 154:83 155:83 156:83 157:84 158:84 159:85 160:86 161:87 162:88 163:89 164:89 165:89 166:90 167:91 168:92 169:93 170:94 171:95 172:96 173:97 174:98 175:99 176:99 177:100 178:100 179:100 180:101 181:101 182:102 183:103 184:104 185:105 186:106 187:107 188:108 189:109 190:110 191:110 192:111 193:112 194:112 195:112 196:112 197:113 198:114 199:115 200:116 201:117 202:118 203:119 204:120 205:121 206:121 207:121 208:122 209:122 210:122 211:123 212:123 I0421 11:11:29.248157 47076539613632 squad_lib.py:379] token_is_max_context: 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True 204:True 205:True 206:True 207:True 208:True 209:True 210:True 211:True 212:True I0421 11:11:29.248379 47076539613632 squad_lib.py:381] input_ids: 101 1191 187 27085 183 15447 16179 1127 1215 117 1184 1156 7688 7329 1851 1138 1151 1270 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.248602 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.248815 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.254668 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.254798 47076539613632 squad_lib.py:367] unique_id: 1000000015 I0421 11:11:29.254912 47076539613632 squad_lib.py:368] example_index: 15 I0421 11:11:29.255022 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.255229 47076539613632 squad_lib.py:371] tokens: [CLS] super bowl 50 decided the n ##f ##l champion for what season ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.255420 47076539613632 squad_lib.py:374] token_to_orig_map: 15:0 16:1 17:2 18:3 19:4 20:5 21:5 22:5 23:6 24:7 25:8 26:9 27:10 28:11 29:12 30:13 31:14 32:15 33:16 34:17 35:17 36:17 37:17 38:17 39:18 40:19 41:20 42:21 43:21 44:22 45:23 46:23 47:23 48:24 49:25 50:26 51:26 52:26 53:26 54:26 55:27 56:28 57:28 58:29 59:29 60:29 61:30 62:31 63:32 64:33 65:34 66:35 67:35 68:35 69:35 70:35 71:36 72:37 73:37 74:37 75:38 76:38 77:38 78:39 79:39 80:39 81:40 82:41 83:42 84:43 85:44 86:45 87:46 88:46 89:47 90:48 91:49 92:50 93:51 94:52 95:52 96:52 97:52 98:53 99:53 100:54 101:54 102:55 103:56 104:56 105:56 106:56 107:57 108:58 109:59 110:60 111:60 112:61 113:61 114:61 115:61 116:62 117:63 118:64 119:65 120:65 121:66 122:66 123:66 124:66 125:67 126:67 127:67 128:67 129:67 130:67 131:68 132:69 133:70 134:71 135:72 136:73 137:74 138:74 139:75 140:76 141:77 142:78 143:79 144:79 145:80 146:80 147:81 148:82 149:83 150:83 151:83 152:84 153:84 154:85 155:86 156:87 157:88 158:89 159:89 160:89 161:90 162:91 163:92 164:93 165:94 166:95 167:96 168:97 169:98 170:99 171:99 172:100 173:100 174:100 175:101 176:101 177:102 178:103 179:104 180:105 181:106 182:107 183:108 184:109 185:110 186:110 187:111 188:112 189:112 190:112 191:112 192:113 193:114 194:115 195:116 196:117 197:118 198:119 199:120 200:121 201:121 202:121 203:122 204:122 205:122 206:123 207:123 I0421 11:11:29.255608 47076539613632 squad_lib.py:379] token_is_max_context: 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True 204:True 205:True 206:True 207:True I0421 11:11:29.255824 47076539613632 squad_lib.py:381] input_ids: 101 7688 7329 1851 1879 1103 183 2087 1233 3628 1111 1184 1265 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.256043 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.256251 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.262142 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.262271 47076539613632 squad_lib.py:367] unique_id: 1000000016 I0421 11:11:29.262397 47076539613632 squad_lib.py:368] example_index: 16 I0421 11:11:29.262506 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.262727 47076539613632 squad_lib.py:371] tokens: [CLS] what year did the den ##ver br ##on ##cos secure a super bowl title for the third time ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.262908 47076539613632 squad_lib.py:374] token_to_orig_map: 21:0 22:1 23:2 24:3 25:4 26:5 27:5 28:5 29:6 30:7 31:8 32:9 33:10 34:11 35:12 36:13 37:14 38:15 39:16 40:17 41:17 42:17 43:17 44:17 45:18 46:19 47:20 48:21 49:21 50:22 51:23 52:23 53:23 54:24 55:25 56:26 57:26 58:26 59:26 60:26 61:27 62:28 63:28 64:29 65:29 66:29 67:30 68:31 69:32 70:33 71:34 72:35 73:35 74:35 75:35 76:35 77:36 78:37 79:37 80:37 81:38 82:38 83:38 84:39 85:39 86:39 87:40 88:41 89:42 90:43 91:44 92:45 93:46 94:46 95:47 96:48 97:49 98:50 99:51 100:52 101:52 102:52 103:52 104:53 105:53 106:54 107:54 108:55 109:56 110:56 111:56 112:56 113:57 114:58 115:59 116:60 117:60 118:61 119:61 120:61 121:61 122:62 123:63 124:64 125:65 126:65 127:66 128:66 129:66 130:66 131:67 132:67 133:67 134:67 135:67 136:67 137:68 138:69 139:70 140:71 141:72 142:73 143:74 144:74 145:75 146:76 147:77 148:78 149:79 150:79 151:80 152:80 153:81 154:82 155:83 156:83 157:83 158:84 159:84 160:85 161:86 162:87 163:88 164:89 165:89 166:89 167:90 168:91 169:92 170:93 171:94 172:95 173:96 174:97 175:98 176:99 177:99 178:100 179:100 180:100 181:101 182:101 183:102 184:103 185:104 186:105 187:106 188:107 189:108 190:109 191:110 192:110 193:111 194:112 195:112 196:112 197:112 198:113 199:114 200:115 201:116 202:117 203:118 204:119 205:120 206:121 207:121 208:121 209:122 210:122 211:122 212:123 213:123 I0421 11:11:29.263093 47076539613632 squad_lib.py:379] token_is_max_context: 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True 204:True 205:True 206:True 207:True 208:True 209:True 210:True 211:True 212:True 213:True I0421 11:11:29.263322 47076539613632 squad_lib.py:381] input_ids: 101 1184 1214 1225 1103 10552 4121 9304 1320 13538 5343 170 7688 7329 1641 1111 1103 1503 1159 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.263537 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.263753 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.269463 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.269592 47076539613632 squad_lib.py:367] unique_id: 1000000017 I0421 11:11:29.269708 47076539613632 squad_lib.py:368] example_index: 17 I0421 11:11:29.269818 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.270023 47076539613632 squad_lib.py:371] tokens: [CLS] what city did super bowl 50 take place in ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.270205 47076539613632 squad_lib.py:374] token_to_orig_map: 12:0 13:1 14:2 15:3 16:4 17:5 18:5 19:5 20:6 21:7 22:8 23:9 24:10 25:11 26:12 27:13 28:14 29:15 30:16 31:17 32:17 33:17 34:17 35:17 36:18 37:19 38:20 39:21 40:21 41:22 42:23 43:23 44:23 45:24 46:25 47:26 48:26 49:26 50:26 51:26 52:27 53:28 54:28 55:29 56:29 57:29 58:30 59:31 60:32 61:33 62:34 63:35 64:35 65:35 66:35 67:35 68:36 69:37 70:37 71:37 72:38 73:38 74:38 75:39 76:39 77:39 78:40 79:41 80:42 81:43 82:44 83:45 84:46 85:46 86:47 87:48 88:49 89:50 90:51 91:52 92:52 93:52 94:52 95:53 96:53 97:54 98:54 99:55 100:56 101:56 102:56 103:56 104:57 105:58 106:59 107:60 108:60 109:61 110:61 111:61 112:61 113:62 114:63 115:64 116:65 117:65 118:66 119:66 120:66 121:66 122:67 123:67 124:67 125:67 126:67 127:67 128:68 129:69 130:70 131:71 132:72 133:73 134:74 135:74 136:75 137:76 138:77 139:78 140:79 141:79 142:80 143:80 144:81 145:82 146:83 147:83 148:83 149:84 150:84 151:85 152:86 153:87 154:88 155:89 156:89 157:89 158:90 159:91 160:92 161:93 162:94 163:95 164:96 165:97 166:98 167:99 168:99 169:100 170:100 171:100 172:101 173:101 174:102 175:103 176:104 177:105 178:106 179:107 180:108 181:109 182:110 183:110 184:111 185:112 186:112 187:112 188:112 189:113 190:114 191:115 192:116 193:117 194:118 195:119 196:120 197:121 198:121 199:121 200:122 201:122 202:122 203:123 204:123 I0421 11:11:29.270397 47076539613632 squad_lib.py:379] token_is_max_context: 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True 204:True I0421 11:11:29.270618 47076539613632 squad_lib.py:381] input_ids: 101 1184 1331 1225 7688 7329 1851 1321 1282 1107 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.270835 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.271047 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.276791 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.276922 47076539613632 squad_lib.py:367] unique_id: 1000000018 I0421 11:11:29.277035 47076539613632 squad_lib.py:368] example_index: 18 I0421 11:11:29.277144 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.277358 47076539613632 squad_lib.py:371] tokens: [CLS] what stadium did super bowl 50 take place in ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.277544 47076539613632 squad_lib.py:374] token_to_orig_map: 12:0 13:1 14:2 15:3 16:4 17:5 18:5 19:5 20:6 21:7 22:8 23:9 24:10 25:11 26:12 27:13 28:14 29:15 30:16 31:17 32:17 33:17 34:17 35:17 36:18 37:19 38:20 39:21 40:21 41:22 42:23 43:23 44:23 45:24 46:25 47:26 48:26 49:26 50:26 51:26 52:27 53:28 54:28 55:29 56:29 57:29 58:30 59:31 60:32 61:33 62:34 63:35 64:35 65:35 66:35 67:35 68:36 69:37 70:37 71:37 72:38 73:38 74:38 75:39 76:39 77:39 78:40 79:41 80:42 81:43 82:44 83:45 84:46 85:46 86:47 87:48 88:49 89:50 90:51 91:52 92:52 93:52 94:52 95:53 96:53 97:54 98:54 99:55 100:56 101:56 102:56 103:56 104:57 105:58 106:59 107:60 108:60 109:61 110:61 111:61 112:61 113:62 114:63 115:64 116:65 117:65 118:66 119:66 120:66 121:66 122:67 123:67 124:67 125:67 126:67 127:67 128:68 129:69 130:70 131:71 132:72 133:73 134:74 135:74 136:75 137:76 138:77 139:78 140:79 141:79 142:80 143:80 144:81 145:82 146:83 147:83 148:83 149:84 150:84 151:85 152:86 153:87 154:88 155:89 156:89 157:89 158:90 159:91 160:92 161:93 162:94 163:95 164:96 165:97 166:98 167:99 168:99 169:100 170:100 171:100 172:101 173:101 174:102 175:103 176:104 177:105 178:106 179:107 180:108 181:109 182:110 183:110 184:111 185:112 186:112 187:112 188:112 189:113 190:114 191:115 192:116 193:117 194:118 195:119 196:120 197:121 198:121 199:121 200:122 201:122 202:122 203:123 204:123 I0421 11:11:29.277722 47076539613632 squad_lib.py:379] token_is_max_context: 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True 204:True I0421 11:11:29.277944 47076539613632 squad_lib.py:381] input_ids: 101 1184 4706 1225 7688 7329 1851 1321 1282 1107 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.278156 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.278375 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.284075 47076539613632 squad_lib.py:366] *** Example *** I0421 11:11:29.284207 47076539613632 squad_lib.py:367] unique_id: 1000000019 I0421 11:11:29.284345 47076539613632 squad_lib.py:368] example_index: 19 I0421 11:11:29.284456 47076539613632 squad_lib.py:369] doc_span_index: 0 I0421 11:11:29.284663 47076539613632 squad_lib.py:371] tokens: [CLS] what was the final score of super bowl 50 ? [SEP] super bowl 50 was an am ##eric ##an football game to determine the champion of the national football league ( n ##f ##l ) for the 2015 season . the am ##eric ##an football conference ( a ##f ##c ) champion den ##ver br ##on ##cos defeated the national football conference ( n ##f ##c ) champion car ##olin ##a pan ##ther ##s 24 – 10 to earn their third super bowl title . the game was played on f ##eb ##ru ##ary 7 , 2016 , at le ##vi ' s stadium in the sa ##n f ##ran ##cis ##co bay area at sa ##nta c ##lar ##a , ca ##li ##fo ##rn ##ia . as this was the 50th super bowl , the league emphasized the " golden anniversary " with various gold - themed initiatives , as well as temporarily su ##sp ##ending the tradition of naming each super bowl game with r ##oman n ##ume ##rals ( under which the game would have been known as " super bowl l " ) , so that the logo could prominently feature the a ##rab ##ic n ##ume ##rals 50 . [SEP] I0421 11:11:29.284847 47076539613632 squad_lib.py:374] token_to_orig_map: 12:0 13:1 14:2 15:3 16:4 17:5 18:5 19:5 20:6 21:7 22:8 23:9 24:10 25:11 26:12 27:13 28:14 29:15 30:16 31:17 32:17 33:17 34:17 35:17 36:18 37:19 38:20 39:21 40:21 41:22 42:23 43:23 44:23 45:24 46:25 47:26 48:26 49:26 50:26 51:26 52:27 53:28 54:28 55:29 56:29 57:29 58:30 59:31 60:32 61:33 62:34 63:35 64:35 65:35 66:35 67:35 68:36 69:37 70:37 71:37 72:38 73:38 74:38 75:39 76:39 77:39 78:40 79:41 80:42 81:43 82:44 83:45 84:46 85:46 86:47 87:48 88:49 89:50 90:51 91:52 92:52 93:52 94:52 95:53 96:53 97:54 98:54 99:55 100:56 101:56 102:56 103:56 104:57 105:58 106:59 107:60 108:60 109:61 110:61 111:61 112:61 113:62 114:63 115:64 116:65 117:65 118:66 119:66 120:66 121:66 122:67 123:67 124:67 125:67 126:67 127:67 128:68 129:69 130:70 131:71 132:72 133:73 134:74 135:74 136:75 137:76 138:77 139:78 140:79 141:79 142:80 143:80 144:81 145:82 146:83 147:83 148:83 149:84 150:84 151:85 152:86 153:87 154:88 155:89 156:89 157:89 158:90 159:91 160:92 161:93 162:94 163:95 164:96 165:97 166:98 167:99 168:99 169:100 170:100 171:100 172:101 173:101 174:102 175:103 176:104 177:105 178:106 179:107 180:108 181:109 182:110 183:110 184:111 185:112 186:112 187:112 188:112 189:113 190:114 191:115 192:116 193:117 194:118 195:119 196:120 197:121 198:121 199:121 200:122 201:122 202:122 203:123 204:123 I0421 11:11:29.285025 47076539613632 squad_lib.py:379] token_is_max_context: 12:True 13:True 14:True 15:True 16:True 17:True 18:True 19:True 20:True 21:True 22:True 23:True 24:True 25:True 26:True 27:True 28:True 29:True 30:True 31:True 32:True 33:True 34:True 35:True 36:True 37:True 38:True 39:True 40:True 41:True 42:True 43:True 44:True 45:True 46:True 47:True 48:True 49:True 50:True 51:True 52:True 53:True 54:True 55:True 56:True 57:True 58:True 59:True 60:True 61:True 62:True 63:True 64:True 65:True 66:True 67:True 68:True 69:True 70:True 71:True 72:True 73:True 74:True 75:True 76:True 77:True 78:True 79:True 80:True 81:True 82:True 83:True 84:True 85:True 86:True 87:True 88:True 89:True 90:True 91:True 92:True 93:True 94:True 95:True 96:True 97:True 98:True 99:True 100:True 101:True 102:True 103:True 104:True 105:True 106:True 107:True 108:True 109:True 110:True 111:True 112:True 113:True 114:True 115:True 116:True 117:True 118:True 119:True 120:True 121:True 122:True 123:True 124:True 125:True 126:True 127:True 128:True 129:True 130:True 131:True 132:True 133:True 134:True 135:True 136:True 137:True 138:True 139:True 140:True 141:True 142:True 143:True 144:True 145:True 146:True 147:True 148:True 149:True 150:True 151:True 152:True 153:True 154:True 155:True 156:True 157:True 158:True 159:True 160:True 161:True 162:True 163:True 164:True 165:True 166:True 167:True 168:True 169:True 170:True 171:True 172:True 173:True 174:True 175:True 176:True 177:True 178:True 179:True 180:True 181:True 182:True 183:True 184:True 185:True 186:True 187:True 188:True 189:True 190:True 191:True 192:True 193:True 194:True 195:True 196:True 197:True 198:True 199:True 200:True 201:True 202:True 203:True 204:True I0421 11:11:29.285244 47076539613632 squad_lib.py:381] input_ids: 101 1184 1108 1103 1509 2794 1104 7688 7329 1851 136 102 7688 7329 1851 1108 1126 1821 26237 1389 1709 1342 1106 4959 1103 3628 1104 1103 1569 1709 2074 113 183 2087 1233 114 1111 1103 1410 1265 119 1103 1821 26237 1389 1709 3511 113 170 2087 1665 114 3628 10552 4121 9304 1320 13538 2378 1103 1569 1709 3511 113 183 2087 1665 114 3628 1610 27719 1161 13316 8420 1116 1572 782 1275 1106 7379 1147 1503 7688 7329 1641 119 1103 1342 1108 1307 1113 175 15581 5082 3113 128 117 1446 117 1120 5837 5086 112 188 4706 1107 1103 21718 1179 175 4047 21349 2528 5952 1298 1120 21718 13130 172 5815 1161 117 11019 2646 14467 4558 1465 119 1112 1142 1108 1103 13163 7688 7329 117 1103 2074 13463 1103 107 5404 5453 107 1114 1672 2284 118 12005 11751 117 1112 1218 1112 7818 28117 20080 16264 1103 3904 1104 10505 1296 7688 7329 1342 1114 187 27085 183 15447 16179 113 1223 1134 1103 1342 1156 1138 1151 1227 1112 107 7688 7329 181 107 114 117 1177 1115 1103 7998 1180 15199 2672 1103 170 17952 1596 183 15447 16179 1851 119 102 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.285465 47076539613632 squad_lib.py:382] input_mask: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:11:29.285675 47076539613632 squad_lib.py:383] segment_ids: 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I0421 11:12:29.524208 47076539613632 squad_lib.py:420] Adding padding examples to make sure no partial batch. I0421 11:12:29.524480 47076539613632 squad_lib.py:421] Adds 2 padding examples for inference. I0421 11:12:29.538662 47076539613632 run_squad_helper.py:341] ***** Running predictions ***** I0421 11:12:29.538791 47076539613632 run_squad_helper.py:342] Num orig examples = 10570 I0421 11:12:29.538907 47076539613632 run_squad_helper.py:343] Num split examples = 10970 I0421 11:12:29.539023 47076539613632 run_squad_helper.py:344] Batch size = 4 2021-04-21 11:12:30.009568: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 11:12:30.016746: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 11:12:30.021664: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 11:12:30.153649: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 11:12:30.177027: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 11:12:30.182349: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 11:12:30.187927: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 11:12:30.193027: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 11:12:30.198579: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 11:12:30.203624: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 11:12:30.209108: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. 2021-04-21 11:12:38.476456: I tensorflow/core/common_runtime/gpu_fusion_pass.cc:508] ROCm Fusion is enabled. I0421 11:12:40.405827 47076539613632 run_squad_helper.py:223] Made predictions for 100 records. I0421 11:12:41.230522 47076539613632 run_squad_helper.py:223] Made predictions for 200 records. I0421 11:12:42.061466 47076539613632 run_squad_helper.py:223] Made predictions for 300 records. I0421 11:12:42.879084 47076539613632 run_squad_helper.py:223] Made predictions for 400 records. I0421 11:12:43.701066 47076539613632 run_squad_helper.py:223] Made predictions for 500 records. I0421 11:12:44.522982 47076539613632 run_squad_helper.py:223] Made predictions for 600 records. I0421 11:12:45.343106 47076539613632 run_squad_helper.py:223] Made predictions for 700 records. I0421 11:12:46.168502 47076539613632 run_squad_helper.py:223] Made predictions for 800 records. I0421 11:12:46.999982 47076539613632 run_squad_helper.py:223] Made predictions for 900 records. I0421 11:12:47.825019 47076539613632 run_squad_helper.py:223] Made predictions for 1000 records. I0421 11:12:48.656949 47076539613632 run_squad_helper.py:223] Made predictions for 1100 records. I0421 11:12:49.484330 47076539613632 run_squad_helper.py:223] Made predictions for 1200 records. I0421 11:12:50.308649 47076539613632 run_squad_helper.py:223] Made predictions for 1300 records. I0421 11:12:51.132554 47076539613632 run_squad_helper.py:223] Made predictions for 1400 records. I0421 11:12:51.963085 47076539613632 run_squad_helper.py:223] Made predictions for 1500 records. I0421 11:12:52.791450 47076539613632 run_squad_helper.py:223] Made predictions for 1600 records. I0421 11:12:53.617588 47076539613632 run_squad_helper.py:223] Made predictions for 1700 records. I0421 11:12:54.443922 47076539613632 run_squad_helper.py:223] Made predictions for 1800 records. I0421 11:12:55.269884 47076539613632 run_squad_helper.py:223] Made predictions for 1900 records. I0421 11:12:56.099067 47076539613632 run_squad_helper.py:223] Made predictions for 2000 records. I0421 11:12:56.932780 47076539613632 run_squad_helper.py:223] Made predictions for 2100 records. I0421 11:12:57.765176 47076539613632 run_squad_helper.py:223] Made predictions for 2200 records. I0421 11:12:58.605224 47076539613632 run_squad_helper.py:223] Made predictions for 2300 records. I0421 11:12:59.451265 47076539613632 run_squad_helper.py:223] Made predictions for 2400 records. I0421 11:13:00.284780 47076539613632 run_squad_helper.py:223] Made predictions for 2500 records. I0421 11:13:01.108484 47076539613632 run_squad_helper.py:223] Made predictions for 2600 records. I0421 11:13:01.939439 47076539613632 run_squad_helper.py:223] Made predictions for 2700 records. I0421 11:13:02.769769 47076539613632 run_squad_helper.py:223] Made predictions for 2800 records. I0421 11:13:03.598085 47076539613632 run_squad_helper.py:223] Made predictions for 2900 records. I0421 11:13:04.426424 47076539613632 run_squad_helper.py:223] Made predictions for 3000 records. I0421 11:13:05.267606 47076539613632 run_squad_helper.py:223] Made predictions for 3100 records. I0421 11:13:06.097973 47076539613632 run_squad_helper.py:223] Made predictions for 3200 records. I0421 11:13:06.933673 47076539613632 run_squad_helper.py:223] Made predictions for 3300 records. I0421 11:13:07.760985 47076539613632 run_squad_helper.py:223] Made predictions for 3400 records. I0421 11:13:08.589337 47076539613632 run_squad_helper.py:223] Made predictions for 3500 records. I0421 11:13:09.418296 47076539613632 run_squad_helper.py:223] Made predictions for 3600 records. I0421 11:13:10.242460 47076539613632 run_squad_helper.py:223] Made predictions for 3700 records. I0421 11:13:11.067596 47076539613632 run_squad_helper.py:223] Made predictions for 3800 records. I0421 11:13:11.892354 47076539613632 run_squad_helper.py:223] Made predictions for 3900 records. I0421 11:13:12.714057 47076539613632 run_squad_helper.py:223] Made predictions for 4000 records. I0421 11:13:13.540582 47076539613632 run_squad_helper.py:223] Made predictions for 4100 records. I0421 11:13:14.367324 47076539613632 run_squad_helper.py:223] Made predictions for 4200 records. I0421 11:13:15.196892 47076539613632 run_squad_helper.py:223] Made predictions for 4300 records. I0421 11:13:16.026397 47076539613632 run_squad_helper.py:223] Made predictions for 4400 records. I0421 11:13:16.854767 47076539613632 run_squad_helper.py:223] Made predictions for 4500 records. I0421 11:13:17.685170 47076539613632 run_squad_helper.py:223] Made predictions for 4600 records. I0421 11:13:18.521788 47076539613632 run_squad_helper.py:223] Made predictions for 4700 records. I0421 11:13:19.346223 47076539613632 run_squad_helper.py:223] Made predictions for 4800 records. I0421 11:13:20.173502 47076539613632 run_squad_helper.py:223] Made predictions for 4900 records. I0421 11:13:20.999221 47076539613632 run_squad_helper.py:223] Made predictions for 5000 records. I0421 11:13:21.826658 47076539613632 run_squad_helper.py:223] Made predictions for 5100 records. I0421 11:13:22.659233 47076539613632 run_squad_helper.py:223] Made predictions for 5200 records. I0421 11:13:23.494912 47076539613632 run_squad_helper.py:223] Made predictions for 5300 records. I0421 11:13:24.315006 47076539613632 run_squad_helper.py:223] Made predictions for 5400 records. I0421 11:13:25.139277 47076539613632 run_squad_helper.py:223] Made predictions for 5500 records. I0421 11:13:25.965684 47076539613632 run_squad_helper.py:223] Made predictions for 5600 records. I0421 11:13:26.798784 47076539613632 run_squad_helper.py:223] Made predictions for 5700 records. I0421 11:13:27.624611 47076539613632 run_squad_helper.py:223] Made predictions for 5800 records. I0421 11:13:28.451828 47076539613632 run_squad_helper.py:223] Made predictions for 5900 records. I0421 11:13:29.279164 47076539613632 run_squad_helper.py:223] Made predictions for 6000 records. I0421 11:13:30.110749 47076539613632 run_squad_helper.py:223] Made predictions for 6100 records. I0421 11:13:30.938165 47076539613632 run_squad_helper.py:223] Made predictions for 6200 records. I0421 11:13:31.769063 47076539613632 run_squad_helper.py:223] Made predictions for 6300 records. I0421 11:13:32.591205 47076539613632 run_squad_helper.py:223] Made predictions for 6400 records. I0421 11:13:33.418792 47076539613632 run_squad_helper.py:223] Made predictions for 6500 records. I0421 11:13:34.244592 47076539613632 run_squad_helper.py:223] Made predictions for 6600 records. I0421 11:13:35.072892 47076539613632 run_squad_helper.py:223] Made predictions for 6700 records. I0421 11:13:35.896670 47076539613632 run_squad_helper.py:223] Made predictions for 6800 records. I0421 11:13:36.723522 47076539613632 run_squad_helper.py:223] Made predictions for 6900 records. I0421 11:13:37.558058 47076539613632 run_squad_helper.py:223] Made predictions for 7000 records. I0421 11:13:38.385153 47076539613632 run_squad_helper.py:223] Made predictions for 7100 records. I0421 11:13:39.212961 47076539613632 run_squad_helper.py:223] Made predictions for 7200 records. I0421 11:13:40.041708 47076539613632 run_squad_helper.py:223] Made predictions for 7300 records. I0421 11:13:40.877595 47076539613632 run_squad_helper.py:223] Made predictions for 7400 records. I0421 11:13:41.706880 47076539613632 run_squad_helper.py:223] Made predictions for 7500 records. I0421 11:13:42.537755 47076539613632 run_squad_helper.py:223] Made predictions for 7600 records. I0421 11:13:43.373473 47076539613632 run_squad_helper.py:223] Made predictions for 7700 records. I0421 11:13:44.207040 47076539613632 run_squad_helper.py:223] Made predictions for 7800 records. I0421 11:13:45.035125 47076539613632 run_squad_helper.py:223] Made predictions for 7900 records. I0421 11:13:45.871198 47076539613632 run_squad_helper.py:223] Made predictions for 8000 records. I0421 11:13:46.698966 47076539613632 run_squad_helper.py:223] Made predictions for 8100 records. I0421 11:13:47.522164 47076539613632 run_squad_helper.py:223] Made predictions for 8200 records. I0421 11:13:48.348444 47076539613632 run_squad_helper.py:223] Made predictions for 8300 records. I0421 11:13:49.168892 47076539613632 run_squad_helper.py:223] Made predictions for 8400 records. I0421 11:13:50.002065 47076539613632 run_squad_helper.py:223] Made predictions for 8500 records. I0421 11:13:50.832612 47076539613632 run_squad_helper.py:223] Made predictions for 8600 records. I0421 11:13:51.662652 47076539613632 run_squad_helper.py:223] Made predictions for 8700 records. I0421 11:13:52.485176 47076539613632 run_squad_helper.py:223] Made predictions for 8800 records. I0421 11:13:53.316275 47076539613632 run_squad_helper.py:223] Made predictions for 8900 records. I0421 11:13:54.142509 47076539613632 run_squad_helper.py:223] Made predictions for 9000 records. I0421 11:13:54.966708 47076539613632 run_squad_helper.py:223] Made predictions for 9100 records. I0421 11:13:55.790188 47076539613632 run_squad_helper.py:223] Made predictions for 9200 records. I0421 11:13:56.615890 47076539613632 run_squad_helper.py:223] Made predictions for 9300 records. I0421 11:13:57.445017 47076539613632 run_squad_helper.py:223] Made predictions for 9400 records. I0421 11:13:58.268332 47076539613632 run_squad_helper.py:223] Made predictions for 9500 records. I0421 11:13:59.093684 47076539613632 run_squad_helper.py:223] Made predictions for 9600 records. I0421 11:13:59.918689 47076539613632 run_squad_helper.py:223] Made predictions for 9700 records. I0421 11:14:00.746421 47076539613632 run_squad_helper.py:223] Made predictions for 9800 records. I0421 11:14:01.570352 47076539613632 run_squad_helper.py:223] Made predictions for 9900 records. I0421 11:14:02.394983 47076539613632 run_squad_helper.py:223] Made predictions for 10000 records. I0421 11:14:03.235595 47076539613632 run_squad_helper.py:223] Made predictions for 10100 records. I0421 11:14:04.057467 47076539613632 run_squad_helper.py:223] Made predictions for 10200 records. I0421 11:14:04.891264 47076539613632 run_squad_helper.py:223] Made predictions for 10300 records. I0421 11:14:05.720426 47076539613632 run_squad_helper.py:223] Made predictions for 10400 records. I0421 11:14:06.553259 47076539613632 run_squad_helper.py:223] Made predictions for 10500 records. I0421 11:14:07.380370 47076539613632 run_squad_helper.py:223] Made predictions for 10600 records. I0421 11:14:08.208481 47076539613632 run_squad_helper.py:223] Made predictions for 10700 records. I0421 11:14:09.038475 47076539613632 run_squad_helper.py:223] Made predictions for 10800 records. I0421 11:14:09.867308 47076539613632 run_squad_helper.py:223] Made predictions for 10900 records. I0421 11:15:08.630571 47076539613632 run_squad_helper.py:374] Writing predictions to: /public/home/xuanbaby/DL-TensorFlow/models_r2.3.0/official/nlp/bert/model_squad_v2/predictions.json I0421 11:15:08.630922 47076539613632 run_squad_helper.py:375] Writing nbest to: /public/home/xuanbaby/DL-TensorFlow/models_r2.3.0/official/nlp/bert/model_squad_v2/nbest_predictions.json I0421 11:15:13.943372 47076539613632 run_squad_xuan.py:147] SQuAD eval F1-score: 0.853577 PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU') memory growth: True PhysicalDevice(name='/physical_device:GPU:1', device_type='GPU') memory growth: True PhysicalDevice(name='/physical_device:GPU:2', device_type='GPU') memory growth: True PhysicalDevice(name='/physical_device:GPU:3', device_type='GPU') memory growth: True