Initial commit

debug/workerlog.3

+======================= Modified FLAGS detected =======================
+FLAGS(name='FLAGS_selected_gpus', current_value='3', default_value='')
+=======================================================================
+I0619 22:24:47.332235 27371 tcp_utils.cc:130] Successfully connected to 10.16.4.9:43062
+I0619 22:24:47.373123 27371 process_group_nccl.cc:120] ProcessGroupNCCL pg_timeout_ 1800000
+W0619 22:24:47.384255 27371 gpu_resources.cc:119] Please NOTE: device: 3, GPU Compute Capability: 90.2, Driver API Version: 50724.6, Runtime API Version: 50724.6

debug/workerlog.4

+======================= Modified FLAGS detected =======================
+FLAGS(name='FLAGS_selected_gpus', current_value='4', default_value='')
+=======================================================================
+I0619 22:24:47.275719 27373 tcp_utils.cc:107] Retry to connect to 10.16.4.9:43062 while the server is not yet listening.
+I0619 22:24:50.275892 27373 tcp_utils.cc:130] Successfully connected to 10.16.4.9:43062
+I0619 22:24:50.318064 27373 process_group_nccl.cc:120] ProcessGroupNCCL pg_timeout_ 1800000
+W0619 22:24:50.329497 27373 gpu_resources.cc:119] Please NOTE: device: 4, GPU Compute Capability: 90.2, Driver API Version: 50724.6, Runtime API Version: 50724.6

debug/workerlog.5

+======================= Modified FLAGS detected =======================
+FLAGS(name='FLAGS_selected_gpus', current_value='5', default_value='')
+=======================================================================
+I0619 22:24:47.278175 27375 tcp_utils.cc:107] Retry to connect to 10.16.4.9:43062 while the server is not yet listening.
+I0619 22:24:50.278354 27375 tcp_utils.cc:130] Successfully connected to 10.16.4.9:43062
+I0619 22:24:50.318130 27375 process_group_nccl.cc:120] ProcessGroupNCCL pg_timeout_ 1800000
+W0619 22:24:50.329583 27375 gpu_resources.cc:119] Please NOTE: device: 5, GPU Compute Capability: 90.2, Driver API Version: 50724.6, Runtime API Version: 50724.6

debug/workerlog.6

+======================= Modified FLAGS detected =======================
+FLAGS(name='FLAGS_selected_gpus', current_value='6', default_value='')
+=======================================================================
+I0619 22:24:47.335712 27378 tcp_utils.cc:130] Successfully connected to 10.16.4.9:43062
+I0619 22:24:47.373050 27378 process_group_nccl.cc:120] ProcessGroupNCCL pg_timeout_ 1800000
+W0619 22:24:47.382917 27378 gpu_resources.cc:119] Please NOTE: device: 6, GPU Compute Capability: 90.2, Driver API Version: 50724.6, Runtime API Version: 50724.6

debug/workerlog.7

+======================= Modified FLAGS detected =======================
+FLAGS(name='FLAGS_selected_gpus', current_value='7', default_value='')
+=======================================================================
+I0619 22:24:47.298169 27381 tcp_utils.cc:130] Successfully connected to 10.16.4.9:43062
+I0619 22:24:47.332026 27381 process_group_nccl.cc:120] ProcessGroupNCCL pg_timeout_ 1800000
+W0619 22:24:47.341987 27381 gpu_resources.cc:119] Please NOTE: device: 7, GPU Compute Capability: 90.2, Driver API Version: 50724.6, Runtime API Version: 50724.6

multi_train.sh

+# recommended paddle.__version__ == 2.0.0
+#python3 -m paddle.distributed.launch --log_dir=./debug/ --gpus '0,1,2,3,4,5,6,7'  tools/train.py -c configs/rec/rec_mv3_none_bilstm_ctc.yml
+wget -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/pretrained/MobileNetV3_large_x0_5_pretrained.pdparams
+
+python3 -m paddle.distributed.launch --log_dir=./debug/ --gpus '0,1,2,3,4,5,6,7'  tools/train.py -c configs/det/det_mv3_db.yml -o Global.epoch_num=50  Architecture.Backbone.scale=1.25 Global.eval_batch_step=[0,40] Train.loader.batch_size_per_card=36 Train.loader.num_workers=2 Eval.loader.num_workers=0  2>&1 | tee ./$(date '+%Y-%m-%dT%H_%M_%S')_dbnet
+

requirements.txt

@@ -4,12 +4,14 @@ imgaug==0.4.0
 pyclipper
 lmdb
 tqdm
-numpy
+numpy==1.23.2 
 visualdl
 rapidfuzz
-opencv-contrib-python==4.4.0.46
+opencv-contrib-python
 cython
 lxml
 premailer
 openpyxl
 attrdict
+pyyaml 
+

single_train.sh

+wget -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/pretrained/MobileNetV3_large_x0_5_pretrained.pdparams
+
+python   tools/train.py -c configs/det/det_mv3_db.yml -o Global.epoch_num=50  Architecture.Backbone.scale=1.25 Global.eval_batch_step=[0,40] Train.loader.batch_size_per_card=36 Train.loader.num_workers=2 Eval.loader.num_workers=0  2>&1 | tee ./$(date '+%Y-%m-%dT%H_%M_%S')_dbnet