YAML clean up

official/vision/beta/projects/yolo/configs/experiments/csp_darknet53.yaml

+runtime:
+  distribution_strategy: 'mirrored'
+  mixed_precision_dtype: 'float32'
+task:
+  model:
+    num_classes: 1001
+    input_size: [256, 256, 3]
+    backbone:
+      type: 'darknet'
+      darknet:
+        model_id: 'cspdarknettiny'
+  losses:
+    l2_weight_decay: 0.0005
+    one_hot: True
+    label_smoothing: 0.1
+  train_data:
+    input_path: 'imagenet-2012-tfrecord/train*'
+    is_training: true
+    global_batch_size: 128
+    dtype: 'float16'
+  validation_data:
+    input_path: 'imagenet-2012-tfrecord/valid*'
+    is_training: true
+    global_batch_size: 128
+    dtype: 'float16'
+    drop_remainder: false
+trainer:
+  train_steps: 1200000 # epochs: 120
+  validation_steps: 400 # size of validation data
+  validation_interval: 10000
+  steps_per_loop: 10000
+  summary_interval: 10000
+  checkpoint_interval: 10000
+  optimizer_config:
+    optimizer:
+      type: 'sgd'
+      sgd:
+        momentum: 0.9
+    learning_rate:
+      type: 'polynomial'
+      polynomial:
+        initial_learning_rate: 0.1
+        end_learning_rate: 0.0001
+        power: 4.0
+        decay_steps: 1190000
+    warmup:
+      type: 'linear'
+      linear:
+        warmup_steps: 1000 #learning rate rises from 0 to 0.1 over 1000 steps

official/vision/beta/projects/yolo/configs/experiments/darknet53.yaml

@@ -4,7 +4,7 @@ runtime:
 task:
   model:
     num_classes: 1001
-    input_size: [224, 224, 3]
+    input_size: [256, 256, 3]
     backbone:
       type: 'darknet'
       darknet:
@@ -20,11 +20,11 @@ task:
   validation_data:
     input_path: 'imagenet-2012-tfrecord/valid*'
     is_training: true
-    global_batch_size: 128 
+    global_batch_size: 128
     dtype: 'float16'
     drop_remainder: false
 trainer:
-  train_steps: 800000 # in the paper
+  train_steps: 800000 # epochs: 80
   validation_steps: 400 # size of validation data
   validation_interval: 10000
   steps_per_loop: 10000
@@ -40,7 +40,7 @@ trainer:
       polynomial:
         initial_learning_rate: 0.1
         end_learning_rate: 0.0001
-        power: 4.0 
+        power: 4.0
         decay_steps: 799000
     warmup:
       type: 'linear'

training_dir/params.yaml

-runtime:
-  all_reduce_alg: null
-  batchnorm_spatial_persistent: false
-  dataset_num_private_threads: null
-  default_shard_dim: -1
-  distribution_strategy: mirrored
-  enable_xla: false
-  gpu_thread_mode: null
-  loss_scale: null
-  mixed_precision_dtype: float32
-  num_cores_per_replica: 1
-  num_gpus: 0
-  num_packs: 1
-  per_gpu_thread_count: 0
-  run_eagerly: false
-  task_index: -1
-  tpu: null
-  worker_hosts: null
-task:
-  gradient_clip_norm: 0.0
-  init_checkpoint: ''
-  logging_dir: null
-  losses:
-    l2_weight_decay: 0.0005
-    label_smoothing: 0.0
-    one_hot: true
-  model:
-    add_head_batch_norm: false
-    backbone:
-      darknet:
-        model_id: cspdarknettiny
-      type: darknet
-    dropout_rate: 0.0
-    input_size: [224, 224, 3]
-    norm_activation:
-      activation: relu
-      norm_epsilon: 0.001
-      norm_momentum: 0.99
-      use_sync_bn: false
-    num_classes: 1001
-  train_data:
-    block_length: 1
-    cache: false
-    cycle_length: 10
-    deterministic: null
-    drop_remainder: true
-    dtype: float16
-    enable_tf_data_service: false
-    global_batch_size: 128
-    input_path: imagenet-2012-tfrecord/train*
-    is_training: true
-    sharding: true
-    shuffle_buffer_size: 10000
-    tf_data_service_address: null
-    tf_data_service_job_name: null
-    tfds_as_supervised: false
-    tfds_data_dir: ''
-    tfds_download: false
-    tfds_name: ''
-    tfds_skip_decoding_feature: ''
-    tfds_split: ''
-  validation_data:
-    block_length: 1
-    cache: false
-    cycle_length: 10
-    deterministic: null
-    drop_remainder: false
-    dtype: float16
-    enable_tf_data_service: false
-    global_batch_size: 128
-    input_path: imagenet-2012-tfrecord/valid*
-    is_training: true
-    sharding: true
-    shuffle_buffer_size: 10000
-    tf_data_service_address: null
-    tf_data_service_job_name: null
-    tfds_as_supervised: false
-    tfds_data_dir: ''
-    tfds_download: false
-    tfds_name: ''
-    tfds_skip_decoding_feature: ''
-    tfds_split: ''
-trainer:
-  allow_tpu_summary: false
-  best_checkpoint_eval_metric: ''
-  best_checkpoint_export_subdir: ''
-  best_checkpoint_metric_comp: higher
-  checkpoint_interval: 10000
-  continuous_eval_timeout: 3600
-  eval_tf_function: true
-  max_to_keep: 5
-  optimizer_config:
-    ema: null
-    learning_rate:
-      polynomial:
-        cycle: false
-        decay_steps: 799000
-        end_learning_rate: 0.0001
-        initial_learning_rate: 0.1
-        name: PolynomialDecay
-        power: 4.0
-      type: polynomial
-    optimizer:
-      sgd:
-        clipnorm: null
-        clipvalue: null
-        decay: 0.0
-        momentum: 0.9
-        name: SGD
-        nesterov: false
-      type: sgd
-    warmup:
-      linear:
-        name: linear
-        warmup_learning_rate: 0
-        warmup_steps: 1000
-      type: linear
-  steps_per_loop: 10000
-  summary_interval: 10000
-  train_steps: 800000
-  train_tf_function: true
-  train_tf_while_loop: true
-  validation_interval: 10000
-  validation_steps: 400