From f1d5d5396e52f0c9497356492ab8fa1eeec1d4f8 Mon Sep 17 00:00:00 2001 From: andyjpaddle Date: Tue, 2 Aug 2022 08:01:37 +0000 Subject: [PATCH 01/25] fix tipc infer log --- test_tipc/test_train_inference_python.sh | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/test_tipc/test_train_inference_python.sh b/test_tipc/test_train_inference_python.sh index 402f636b1b..545cdbba20 100644 --- a/test_tipc/test_train_inference_python.sh +++ b/test_tipc/test_train_inference_python.sh @@ -101,6 +101,7 @@ function func_inference(){ _log_path=$4 _img_dir=$5 _flag_quant=$6 + _gpu=$7 # inference for use_gpu in ${use_gpu_list[*]}; do if [ ${use_gpu} = "False" ] || [ ${use_gpu} = "cpu" ]; then @@ -119,7 +120,7 @@ function func_inference(){ fi # skip when quant model inference but precision is not int8 set_precision=$(func_set_params "${precision_key}" "${precision}") - _save_log_path="${_log_path}/python_infer_cpu_usemkldnn_${use_mkldnn}_threads_${threads}_precision_${precision}_batchsize_${batch_size}.log" + _save_log_path="${_log_path}/python_infer_cpu_gpus_${_gpu}_usemkldnn_${use_mkldnn}_threads_${threads}_precision_${precision}_batchsize_${batch_size}.log" set_infer_data=$(func_set_params "${image_dir_key}" "${_img_dir}") set_benchmark=$(func_set_params "${benchmark_key}" "${benchmark_value}") set_batchsize=$(func_set_params "${batch_size_key}" "${batch_size}") @@ -150,7 +151,7 @@ function func_inference(){ continue fi for batch_size in ${batch_size_list[*]}; do - _save_log_path="${_log_path}/python_infer_gpu_usetrt_${use_trt}_precision_${precision}_batchsize_${batch_size}.log" + _save_log_path="${_log_path}/python_infer_gpu_gpus_${_gpu}_usetrt_${use_trt}_precision_${precision}_batchsize_${batch_size}.log" set_infer_data=$(func_set_params "${image_dir_key}" "${_img_dir}") set_benchmark=$(func_set_params "${benchmark_key}" "${benchmark_value}") set_batchsize=$(func_set_params "${batch_size_key}" "${batch_size}") @@ -184,6 +185,7 @@ if [ ${MODE} = "whole_infer" ]; then # set CUDA_VISIBLE_DEVICES eval $env export Count=0 + gpu=0 IFS="|" infer_run_exports=(${infer_export_list}) infer_quant_flag=(${infer_is_quant}) @@ -205,7 +207,7 @@ if [ ${MODE} = "whole_infer" ]; then fi #run inference is_quant=${infer_quant_flag[Count]} - func_inference "${python}" "${inference_py}" "${save_infer_dir}" "${LOG_PATH}" "${infer_img_dir}" ${is_quant} + func_inference "${python}" "${inference_py}" "${save_infer_dir}" "${LOG_PATH}" "${infer_img_dir}" ${is_quant} "${gpu}" Count=$(($Count + 1)) done else @@ -328,7 +330,7 @@ else else infer_model_dir=${save_infer_path} fi - func_inference "${python}" "${inference_py}" "${infer_model_dir}" "${LOG_PATH}" "${train_infer_img_dir}" "${flag_quant}" + func_inference "${python}" "${inference_py}" "${infer_model_dir}" "${LOG_PATH}" "${train_infer_img_dir}" "${flag_quant}" "${gpu}" eval "unset CUDA_VISIBLE_DEVICES" fi From 160e5e384a2d192a7428d3c53e8ce14c39926ad3 Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Tue, 2 Aug 2022 19:41:29 +0800 Subject: [PATCH 02/25] rename 2.0 to 2_0 --- ppocr/data/imaug/label_ops.py | 9 +- ppstructure/vqa/README.md | 2 +- ppstructure/vqa/README_ch.md | 2 +- ppstructure/vqa/predict_vqa_token_ser.py | 2 +- .../ch_PP-OCRv2_rec_distillation.yml | 2 +- .../ch_PP-OCRv3_rec_distillation.yml | 2 +- .../ch_PP-OCRv3_rec/train_infer_python.txt | 4 +- ..._normal_normal_infer_cpp_linux_gpu_cpu.txt | 2 +- ...rmal_normal_infer_python_linux_gpu_cpu.txt | 2 +- ...nux_gpu_normal_normal_lite_cpp_arm_cpu.txt | 0 ..._normal_normal_lite_cpp_arm_gpu_opencl.txt | 0 ...al_normal_paddle2onnx_python_linux_cpu.txt | 2 +- ...ormal_normal_serving_cpp_linux_gpu_cpu.txt | 2 +- ...al_normal_serving_python_linux_gpu_cpu.txt | 2 +- ..._normal_normal_infer_cpp_linux_gpu_cpu.txt | 2 +- ..._gpu_normal_normal_infer_python_jetson.txt | 2 +- ...nux_gpu_normal_normal_lite_cpp_arm_cpu.txt | 0 ..._normal_normal_lite_cpp_arm_gpu_opencl.txt | 0 ...al_normal_paddle2onnx_python_linux_cpu.txt | 2 +- ...al_normal_serving_python_linux_gpu_cpu.txt | 2 +- .../train_infer_python.txt | 2 +- ...nux_dcu_normal_normal_infer_python_dcu.txt | 0 ...leet_normal_infer_python_linux_gpu_cpu.txt | 2 +- ..._normal_amp_infer_python_linux_gpu_cpu.txt | 2 +- ...cpu_normal_normal_infer_python_mac_cpu.txt | 0 .../train_pact_infer_python.txt | 2 +- .../train_ptq_infer_python.txt | 2 +- ...al_normal_infer_python_windows_cpu_gpu.txt | 0 .../train_infer_python.txt | 2 +- ..._normal_amp_infer_python_linux_gpu_cpu.txt | 2 +- ..._normal_normal_infer_cpp_linux_gpu_cpu.txt | 2 +- ...gpu_normal_normal_infer_python_mac_cpu.txt | 2 +- ...al_normal_infer_python_windows_gpu_cpu.txt | 2 +- ...ormal_normal_serving_cpp_linux_gpu_cpu.txt | 2 +- ...al_normal_serving_python_linux_gpu_cpu.txt | 2 +- ..._normal_normal_infer_cpp_linux_gpu_cpu.txt | 2 +- ...ormal_normal_serving_cpp_linux_gpu_cpu.txt | 2 +- ...al_normal_serving_python_linux_gpu_cpu.txt | 2 +- ..._normal_normal_infer_cpp_linux_gpu_cpu.txt | 2 +- ...al_normal_paddle2onnx_python_linux_cpu.txt | 2 +- ...al_normal_serving_python_linux_gpu_cpu.txt | 2 +- .../train_infer_python.txt | 2 +- ...leet_normal_infer_python_linux_gpu_cpu.txt | 2 +- ..._normal_amp_infer_python_linux_gpu_cpu.txt | 2 +- .../train_pact_infer_python.txt | 2 +- .../train_ptq_infer_python.txt | 2 +- .../rec_chinese_lite_train_v2.0.yml | 0 .../train_infer_python.txt | 2 +- ..._normal_amp_infer_python_linux_gpu_cpu.txt | 2 +- ..._normal_normal_infer_cpp_linux_gpu_cpu.txt | 2 +- ...ormal_normal_serving_cpp_linux_gpu_cpu.txt | 2 +- ...al_normal_serving_python_linux_gpu_cpu.txt | 2 +- .../rec_chinese_lite_train_v2.0.yml | 0 ..._normal_normal_infer_cpp_linux_gpu_cpu.txt | 2 +- ...ormal_normal_serving_cpp_linux_gpu_cpu.txt | 2 +- ...al_normal_serving_python_linux_gpu_cpu.txt | 2 +- .../rec_chinese_lite_train_v2.0.yml | 0 ..._normal_normal_infer_cpp_linux_gpu_cpu.txt | 2 +- ...rmal_normal_infer_python_linux_gpu_cpu.txt | 2 +- ...al_normal_paddle2onnx_python_linux_cpu.txt | 2 +- ...ormal_normal_serving_cpp_linux_gpu_cpu.txt | 2 +- ...al_normal_serving_python_linux_gpu_cpu.txt | 2 +- .../det_r50_vd_db.yml | 0 ..._normal_normal_infer_cpp_linux_gpu_cpu.txt | 2 +- ...al_normal_paddle2onnx_python_linux_cpu.txt | 2 +- ...al_normal_serving_python_linux_gpu_cpu.txt | 2 +- .../train_infer_python.txt | 2 +- ...leet_normal_infer_python_linux_gpu_cpu.txt | 2 +- ..._normal_amp_infer_python_linux_gpu_cpu.txt | 2 +- ..._normal_normal_infer_cpp_linux_gpu_cpu.txt | 2 +- ...al_normal_paddle2onnx_python_linux_cpu.txt | 2 +- ...al_normal_serving_python_linux_gpu_cpu.txt | 2 +- .../rec_icdar15_train.yml | 0 .../train_infer_python.txt | 2 +- ...leet_normal_infer_python_linux_gpu_cpu.txt | 2 +- ..._normal_amp_infer_python_linux_gpu_cpu.txt | 2 +- .../det_mv3_east.yml | 0 .../train_infer_python.txt | 2 +- .../det_mv3_pse.yml | 0 .../train_infer_python.txt | 2 +- .../train_infer_python.txt | 2 +- .../det_r50_vd_dcn_fce_ctw.yml | 0 .../train_infer_python.txt | 2 +- .../det_r50_vd_sast_icdar2015.yml | 0 .../train_infer_python.txt | 2 +- .../det_r50_vd_sast_totaltext.yml | 0 .../train_infer_python.txt | 2 +- .../rec_icdar15_train.yml | 0 .../train_infer_python.txt | 2 +- .../rec_icdar15_train.yml | 0 .../train_infer_python.txt | 2 +- .../rec_mv3_tps_bilstm_att.yml | 0 .../train_infer_python.txt | 2 +- .../rec_icdar15_train.yml | 0 .../train_infer_python.txt | 2 +- .../rec_r32_gaspin_bilstm_att.yml | 2 +- .../train_infer_python.txt | 10 +- .../rec_icdar15_train.yml | 0 .../train_infer_python.txt | 2 +- .../rec_icdar15_train.yml | 0 .../train_infer_python.txt | 2 +- .../rec_r34_vd_tps_bilstm_att.yml | 0 .../train_infer_python.txt | 2 +- .../rec_icdar15_train.yml | 0 .../train_infer_python.txt | 2 +- test_tipc/docs/benchmark_train.md | 2 +- test_tipc/prepare.sh | 120 ++++++++++-------- 107 files changed, 159 insertions(+), 140 deletions(-) rename test_tipc/configs/{ch_ppocr_mobile_v2.0 => ch_ppocr_mobile_v2_0}/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt (94%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0 => ch_ppocr_mobile_v2_0}/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt (94%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0 => ch_ppocr_mobile_v2_0}/model_linux_gpu_normal_normal_lite_cpp_arm_cpu.txt (100%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0 => ch_ppocr_mobile_v2_0}/model_linux_gpu_normal_normal_lite_cpp_arm_gpu_opencl.txt (100%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0 => ch_ppocr_mobile_v2_0}/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt (95%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0 => ch_ppocr_mobile_v2_0}/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt (96%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0 => ch_ppocr_mobile_v2_0}/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt (97%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt (92%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/model_linux_gpu_normal_normal_infer_python_jetson.txt (92%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/model_linux_gpu_normal_normal_lite_cpp_arm_cpu.txt (100%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/model_linux_gpu_normal_normal_lite_cpp_arm_gpu_opencl.txt (100%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt (93%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt (96%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/train_infer_python.txt (98%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/train_linux_dcu_normal_normal_infer_python_dcu.txt (100%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt (97%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt (97%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/train_mac_cpu_normal_normal_infer_python_mac_cpu.txt (100%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/train_pact_infer_python.txt (97%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/train_ptq_infer_python.txt (93%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det => ch_ppocr_mobile_v2_0_det}/train_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt (100%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det_FPGM => ch_ppocr_mobile_v2_0_det_FPGM}/train_infer_python.txt (97%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det_FPGM => ch_ppocr_mobile_v2_0_det_FPGM}/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt (97%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det_KL => ch_ppocr_mobile_v2_0_det_KL}/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt (92%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det_KL => ch_ppocr_mobile_v2_0_det_KL}/model_linux_gpu_normal_normal_infer_python_mac_cpu.txt (89%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det_KL => ch_ppocr_mobile_v2_0_det_KL}/model_linux_gpu_normal_normal_infer_python_windows_gpu_cpu.txt (89%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec_KL => ch_ppocr_mobile_v2_0_det_KL}/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt (95%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det_KL => ch_ppocr_mobile_v2_0_det_KL}/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt (96%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det_PACT => ch_ppocr_mobile_v2_0_det_PACT}/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt (92%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det_PACT => ch_ppocr_mobile_v2_0_det_PACT}/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt (95%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det_PACT => ch_ppocr_mobile_v2_0_det_PACT}/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt (95%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec => ch_ppocr_mobile_v2_0_rec}/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt (93%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec => ch_ppocr_mobile_v2_0_rec}/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt (93%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec => ch_ppocr_mobile_v2_0_rec}/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt (96%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec => ch_ppocr_mobile_v2_0_rec}/train_infer_python.txt (98%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec => ch_ppocr_mobile_v2_0_rec}/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt (97%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec => ch_ppocr_mobile_v2_0_rec}/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt (97%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec => ch_ppocr_mobile_v2_0_rec}/train_pact_infer_python.txt (97%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec => ch_ppocr_mobile_v2_0_rec}/train_ptq_infer_python.txt (94%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec_FPGM => ch_ppocr_mobile_v2_0_rec_FPGM}/rec_chinese_lite_train_v2.0.yml (100%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec_FPGM => ch_ppocr_mobile_v2_0_rec_FPGM}/train_infer_python.txt (97%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec_FPGM => ch_ppocr_mobile_v2_0_rec_FPGM}/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt (97%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec_KL => ch_ppocr_mobile_v2_0_rec_KL}/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt (93%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_det_KL => ch_ppocr_mobile_v2_0_rec_KL}/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt (95%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec_KL => ch_ppocr_mobile_v2_0_rec_KL}/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt (96%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec_KL => ch_ppocr_mobile_v2_0_rec_KL}/rec_chinese_lite_train_v2.0.yml (100%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec_PACT => ch_ppocr_mobile_v2_0_rec_PACT}/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt (92%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec_PACT => ch_ppocr_mobile_v2_0_rec_PACT}/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt (95%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec_PACT => ch_ppocr_mobile_v2_0_rec_PACT}/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt (95%) rename test_tipc/configs/{ch_ppocr_mobile_v2.0_rec_PACT => ch_ppocr_mobile_v2_0_rec_PACT}/rec_chinese_lite_train_v2.0.yml (100%) rename test_tipc/configs/{ch_ppocr_server_v2.0 => ch_ppocr_server_v2_0}/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt (94%) rename test_tipc/configs/{ch_ppocr_server_v2.0 => ch_ppocr_server_v2_0}/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt (94%) rename test_tipc/configs/{ch_ppocr_server_v2.0 => ch_ppocr_server_v2_0}/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt (95%) rename test_tipc/configs/{ch_ppocr_server_v2.0 => ch_ppocr_server_v2_0}/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt (96%) rename test_tipc/configs/{ch_ppocr_server_v2.0 => ch_ppocr_server_v2_0}/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt (97%) rename test_tipc/configs/{ch_ppocr_server_v2.0_det => ch_ppocr_server_v2_0_det}/det_r50_vd_db.yml (100%) rename test_tipc/configs/{ch_ppocr_server_v2.0_det => ch_ppocr_server_v2_0_det}/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt (92%) rename test_tipc/configs/{ch_ppocr_server_v2.0_det => ch_ppocr_server_v2_0_det}/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt (93%) rename test_tipc/configs/{ch_ppocr_server_v2.0_det => ch_ppocr_server_v2_0_det}/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt (96%) rename test_tipc/configs/{ch_ppocr_server_v2.0_det => ch_ppocr_server_v2_0_det}/train_infer_python.txt (98%) rename test_tipc/configs/{ch_ppocr_server_v2.0_det => ch_ppocr_server_v2_0_det}/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt (97%) rename test_tipc/configs/{ch_ppocr_server_v2.0_det => ch_ppocr_server_v2_0_det}/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt (97%) rename test_tipc/configs/{ch_ppocr_server_v2.0_rec => ch_ppocr_server_v2_0_rec}/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt (93%) rename test_tipc/configs/{ch_ppocr_server_v2.0_rec => ch_ppocr_server_v2_0_rec}/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt (93%) rename test_tipc/configs/{ch_ppocr_server_v2.0_rec => ch_ppocr_server_v2_0_rec}/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt (96%) rename test_tipc/configs/{ch_ppocr_server_v2.0_rec => ch_ppocr_server_v2_0_rec}/rec_icdar15_train.yml (100%) rename test_tipc/configs/{ch_ppocr_server_v2.0_rec => ch_ppocr_server_v2_0_rec}/train_infer_python.txt (98%) rename test_tipc/configs/{ch_ppocr_server_v2.0_rec => ch_ppocr_server_v2_0_rec}/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt (97%) rename test_tipc/configs/{ch_ppocr_server_v2.0_rec => ch_ppocr_server_v2_0_rec}/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt (97%) rename test_tipc/configs/{det_mv3_east_v2.0 => det_mv3_east_v2_0}/det_mv3_east.yml (100%) rename test_tipc/configs/{det_mv3_east_v2.0 => det_mv3_east_v2_0}/train_infer_python.txt (98%) rename test_tipc/configs/{det_mv3_pse_v2.0 => det_mv3_pse_v2_0}/det_mv3_pse.yml (100%) rename test_tipc/configs/{det_mv3_pse_v2.0 => det_mv3_pse_v2_0}/train_infer_python.txt (98%) rename test_tipc/configs/{det_r50_db_v2.0 => det_r50_db_v2_0}/train_infer_python.txt (98%) rename test_tipc/configs/{det_r50_dcn_fce_ctw_v2.0 => det_r50_dcn_fce_ctw_v2_0}/det_r50_vd_dcn_fce_ctw.yml (100%) rename test_tipc/configs/{det_r50_dcn_fce_ctw_v2.0 => det_r50_dcn_fce_ctw_v2_0}/train_infer_python.txt (98%) rename test_tipc/configs/{det_r50_vd_sast_icdar15_v2.0 => det_r50_vd_sast_icdar15_v2_0}/det_r50_vd_sast_icdar2015.yml (100%) rename test_tipc/configs/{det_r50_vd_sast_icdar15_v2.0 => det_r50_vd_sast_icdar15_v2_0}/train_infer_python.txt (97%) rename test_tipc/configs/{det_r50_vd_sast_totaltext_v2.0 => det_r50_vd_sast_totaltext_v2_0}/det_r50_vd_sast_totaltext.yml (100%) rename test_tipc/configs/{det_r50_vd_sast_totaltext_v2.0 => det_r50_vd_sast_totaltext_v2_0}/train_infer_python.txt (97%) rename test_tipc/configs/{rec_mv3_none_bilstm_ctc_v2.0 => rec_mv3_none_bilstm_ctc_v2_0}/rec_icdar15_train.yml (100%) rename test_tipc/configs/{rec_mv3_none_bilstm_ctc_v2.0 => rec_mv3_none_bilstm_ctc_v2_0}/train_infer_python.txt (98%) rename test_tipc/configs/{rec_mv3_none_none_ctc_v2.0 => rec_mv3_none_none_ctc_v2_0}/rec_icdar15_train.yml (100%) rename test_tipc/configs/{rec_mv3_none_none_ctc_v2.0 => rec_mv3_none_none_ctc_v2_0}/train_infer_python.txt (97%) rename test_tipc/configs/{rec_mv3_tps_bilstm_att_v2.0 => rec_mv3_tps_bilstm_att_v2_0}/rec_mv3_tps_bilstm_att.yml (100%) rename test_tipc/configs/{rec_mv3_tps_bilstm_att_v2.0 => rec_mv3_tps_bilstm_att_v2_0}/train_infer_python.txt (97%) rename test_tipc/configs/{rec_mv3_tps_bilstm_ctc_v2.0 => rec_mv3_tps_bilstm_ctc_v2_0}/rec_icdar15_train.yml (100%) rename test_tipc/configs/{rec_mv3_tps_bilstm_ctc_v2.0 => rec_mv3_tps_bilstm_ctc_v2_0}/train_infer_python.txt (97%) rename test_tipc/configs/{rec_r34_vd_none_bilstm_ctc_v2.0 => rec_r34_vd_none_bilstm_ctc_v2_0}/rec_icdar15_train.yml (100%) rename test_tipc/configs/{rec_r34_vd_none_bilstm_ctc_v2.0 => rec_r34_vd_none_bilstm_ctc_v2_0}/train_infer_python.txt (97%) rename test_tipc/configs/{rec_r34_vd_none_none_ctc_v2.0 => rec_r34_vd_none_none_ctc_v2_0}/rec_icdar15_train.yml (100%) rename test_tipc/configs/{rec_r34_vd_none_none_ctc_v2.0 => rec_r34_vd_none_none_ctc_v2_0}/train_infer_python.txt (97%) rename test_tipc/configs/{rec_r34_vd_tps_bilstm_att_v2.0 => rec_r34_vd_tps_bilstm_att_v2_0}/rec_r34_vd_tps_bilstm_att.yml (100%) rename test_tipc/configs/{rec_r34_vd_tps_bilstm_att_v2.0 => rec_r34_vd_tps_bilstm_att_v2_0}/train_infer_python.txt (97%) rename test_tipc/configs/{rec_r34_vd_tps_bilstm_ctc_v2.0 => rec_r34_vd_tps_bilstm_ctc_v2_0}/rec_icdar15_train.yml (100%) rename test_tipc/configs/{rec_r34_vd_tps_bilstm_ctc_v2.0 => rec_r34_vd_tps_bilstm_ctc_v2_0}/train_infer_python.txt (97%) diff --git a/ppocr/data/imaug/label_ops.py b/ppocr/data/imaug/label_ops.py index 97539faf23..d84f966201 100644 --- a/ppocr/data/imaug/label_ops.py +++ b/ppocr/data/imaug/label_ops.py @@ -977,7 +977,10 @@ class VQATokenLabelEncode(object): info["bbox"] = self.trans_poly_to_bbox(info["points"]) encode_res = self.tokenizer.encode( - text, pad_to_max_seq_len=False, return_attention_mask=True) + text, + pad_to_max_seq_len=False, + return_attention_mask=True, + return_token_type_ids=True) if not self.add_special_ids: # TODO: use tok.all_special_ids to remove @@ -1217,6 +1220,7 @@ class ABINetLabelEncode(BaseRecLabelEncode): dict_character = [''] + dict_character return dict_character + class SPINLabelEncode(AttnLabelEncode): """ Convert between text-label and text-index """ @@ -1229,6 +1233,7 @@ class SPINLabelEncode(AttnLabelEncode): super(SPINLabelEncode, self).__init__( max_text_length, character_dict_path, use_space_char) self.lower = lower + def add_special_char(self, dict_character): self.beg_str = "sos" self.end_str = "eos" @@ -1248,4 +1253,4 @@ class SPINLabelEncode(AttnLabelEncode): padded_text[:len(target)] = target data['label'] = np.array(padded_text) - return data \ No newline at end of file + return data diff --git a/ppstructure/vqa/README.md b/ppstructure/vqa/README.md index 05635265b5..28b794383b 100644 --- a/ppstructure/vqa/README.md +++ b/ppstructure/vqa/README.md @@ -216,7 +216,7 @@ Use the following command to complete the tandem prediction of `OCR + SER` based ```shell cd ppstructure -CUDA_VISIBLE_DEVICES=0 python3.7 vqa/predict_vqa_token_ser.py --vqa_algorithm=LayoutXLM --ser_model_dir=../output/ser/infer --ser_dict_path=../train_data/XFUND/class_list_xfun.txt --image_dir=docs/vqa/input/zh_val_42.jpg --output=output +CUDA_VISIBLE_DEVICES=0 python3.7 vqa/predict_vqa_token_ser.py --vqa_algorithm=LayoutXLM --ser_model_dir=../output/ser/infer --ser_dict_path=../train_data/XFUND/class_list_xfun.txt --vis_font_path=../doc/fonts/simfang.ttf --image_dir=docs/vqa/input/zh_val_42.jpg --output=output ``` After the prediction is successful, the visualization images and results will be saved in the directory specified by the `output` field diff --git a/ppstructure/vqa/README_ch.md b/ppstructure/vqa/README_ch.md index b421a82d3a..f168110ed9 100644 --- a/ppstructure/vqa/README_ch.md +++ b/ppstructure/vqa/README_ch.md @@ -215,7 +215,7 @@ python3.7 tools/export_model.py -c configs/vqa/ser/layoutxlm.yml -o Architecture ```shell cd ppstructure -CUDA_VISIBLE_DEVICES=0 python3.7 vqa/predict_vqa_token_ser.py --vqa_algorithm=LayoutXLM --ser_model_dir=../output/ser/infer --ser_dict_path=../train_data/XFUND/class_list_xfun.txt --image_dir=docs/vqa/input/zh_val_42.jpg --output=output +CUDA_VISIBLE_DEVICES=0 python3.7 vqa/predict_vqa_token_ser.py --vqa_algorithm=LayoutXLM --ser_model_dir=../output/ser/infer --ser_dict_path=../train_data/XFUND/class_list_xfun.txt --vis_font_path=../doc/fonts/simfang.ttf --image_dir=docs/vqa/input/zh_val_42.jpg --output=output ``` 预测成功后，可视化图片和结果会保存在`output`字段指定的目录下 diff --git a/ppstructure/vqa/predict_vqa_token_ser.py b/ppstructure/vqa/predict_vqa_token_ser.py index de0bbfe72d..3097ebcf16 100644 --- a/ppstructure/vqa/predict_vqa_token_ser.py +++ b/ppstructure/vqa/predict_vqa_token_ser.py @@ -153,7 +153,7 @@ def main(args): img_res = draw_ser_results( image_file, ser_res, - font_path="../doc/fonts/simfang.ttf", ) + font_path=args.vis_font_path, ) img_save_path = os.path.join(args.output, os.path.basename(image_file)) diff --git a/test_tipc/configs/ch_PP-OCRv2_rec/ch_PP-OCRv2_rec_distillation.yml b/test_tipc/configs/ch_PP-OCRv2_rec/ch_PP-OCRv2_rec_distillation.yml index a1497ba8fa..3eb82d42bc 100644 --- a/test_tipc/configs/ch_PP-OCRv2_rec/ch_PP-OCRv2_rec_distillation.yml +++ b/test_tipc/configs/ch_PP-OCRv2_rec/ch_PP-OCRv2_rec_distillation.yml @@ -114,7 +114,7 @@ Train: name: SimpleDataSet data_dir: ./train_data/ic15_data/ label_file_list: - - ./train_data/ic15_data/rec_gt_train4w.txt + - ./train_data/ic15_data/rec_gt_train.txt transforms: - DecodeImage: img_mode: BGR diff --git a/test_tipc/configs/ch_PP-OCRv3_rec/ch_PP-OCRv3_rec_distillation.yml b/test_tipc/configs/ch_PP-OCRv3_rec/ch_PP-OCRv3_rec_distillation.yml index ee884f6687..4c8ba0a6fa 100644 --- a/test_tipc/configs/ch_PP-OCRv3_rec/ch_PP-OCRv3_rec_distillation.yml +++ b/test_tipc/configs/ch_PP-OCRv3_rec/ch_PP-OCRv3_rec_distillation.yml @@ -153,7 +153,7 @@ Train: data_dir: ./train_data/ic15_data/ ext_op_transform_idx: 1 label_file_list: - - ./train_data/ic15_data/rec_gt_train4w.txt + - ./train_data/ic15_data/rec_gt_train.txt transforms: - DecodeImage: img_mode: BGR diff --git a/test_tipc/configs/ch_PP-OCRv3_rec/train_infer_python.txt b/test_tipc/configs/ch_PP-OCRv3_rec/train_infer_python.txt index 420c6592d7..84cceb6682 100644 --- a/test_tipc/configs/ch_PP-OCRv3_rec/train_infer_python.txt +++ b/test_tipc/configs/ch_PP-OCRv3_rec/train_infer_python.txt @@ -52,8 +52,8 @@ null:null ===========================infer_benchmark_params========================== random_infer_input:[{float32,[3,48,320]}] ===========================train_benchmark_params========================== -batch_size:128 +batch_size:64 fp_items:fp32|fp16 epoch:1 --profiler_options:batch_range=[10,20];state=GPU;tracer_option=Default;profile_path=model.profile -flags:FLAGS_eager_delete_tensor_gb=0.0;FLAGS_fraction_of_gpu_memory_to_use=0.98;FLAGS_conv_workspace_size_limit=4096 + diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt similarity index 94% rename from test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt index b42ab9db36..73f1d49855 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================cpp_infer_params=========================== -model_name:ch_ppocr_mobile_v2.0 +model_name:ch_ppocr_mobile_v2_0 use_opencv:True infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ infer_quant:False diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt similarity index 94% rename from test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt index becad991ea..00373b61ee 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================ch_ppocr_mobile_v2.0=========================== -model_name:ch_ppocr_mobile_v2.0 +model_name:ch_ppocr_mobile_v2_0 python:python3.7 infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ infer_export:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_lite_cpp_arm_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_lite_cpp_arm_cpu.txt similarity index 100% rename from test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_lite_cpp_arm_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_lite_cpp_arm_cpu.txt diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_lite_cpp_arm_gpu_opencl.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_lite_cpp_arm_gpu_opencl.txt similarity index 100% rename from test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_lite_cpp_arm_gpu_opencl.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_lite_cpp_arm_gpu_opencl.txt diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt similarity index 95% rename from test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt index 17c2fbbae2..3e01ae5738 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt @@ -1,5 +1,5 @@ ===========================paddle2onnx_params=========================== -model_name:ch_ppocr_mobile_v2.0 +model_name:ch_ppocr_mobile_v2_0 python:python3.7 2onnx: paddle2onnx --det_model_dir:./inference/ch_ppocr_mobile_v2.0_det_infer/ diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt similarity index 96% rename from test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt index d18e9f11fd..305882aa30 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_mobile_v2.0 +model_name:ch_ppocr_mobile_v2_0 python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:./inference/ch_ppocr_mobile_v2.0_det_infer/ diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt index 842c934017..0c366b03de 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_mobile_v2.0 +model_name:ch_ppocr_mobile_v2_0 python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:./inference/ch_ppocr_mobile_v2.0_det_infer/ diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt similarity index 92% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt index 1d1c2ae283..ded332e674 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================cpp_infer_params=========================== -model_name:ch_ppocr_mobile_v2.0_det +model_name:ch_ppocr_mobile_v2_0_det use_opencv:True infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ infer_quant:False diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_infer_python_jetson.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_infer_python_jetson.txt similarity index 92% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_infer_python_jetson.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_infer_python_jetson.txt index 24bb8746ab..5f9dfa5f55 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_infer_python_jetson.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_infer_python_jetson.txt @@ -1,5 +1,5 @@ ===========================infer_params=========================== -model_name:ch_ppocr_mobile_v2.0_det +model_name:ch_ppocr_mobile_v2_0_det python:python infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer infer_export:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_lite_cpp_arm_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_lite_cpp_arm_cpu.txt similarity index 100% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_lite_cpp_arm_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_lite_cpp_arm_cpu.txt diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_lite_cpp_arm_gpu_opencl.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_lite_cpp_arm_gpu_opencl.txt similarity index 100% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_lite_cpp_arm_gpu_opencl.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_lite_cpp_arm_gpu_opencl.txt diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt similarity index 93% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt index 00473d1062..8f36ad4b86 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt @@ -1,5 +1,5 @@ ===========================paddle2onnx_params=========================== -model_name:ch_ppocr_mobile_v2.0_det +model_name:ch_ppocr_mobile_v2_0_det python:python3.7 2onnx: paddle2onnx --det_model_dir:./inference/ch_ppocr_mobile_v2.0_det_infer/ diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt similarity index 96% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt index c9dd5ad920..6dfd7e7bd0 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_mobile_v2.0_det +model_name:ch_ppocr_mobile_v2_0_det python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:./inference/ch_ppocr_mobile_v2.0_det_infer/ diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_infer_python.txt similarity index 98% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_infer_python.txt index 3db816cc08..f3aa9d0f82 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_mobile_v2.0_det +model_name:ch_ppocr_mobile_v2_0_det python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_linux_dcu_normal_normal_infer_python_dcu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_linux_dcu_normal_normal_infer_python_dcu.txt similarity index 100% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_linux_dcu_normal_normal_infer_python_dcu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_linux_dcu_normal_normal_infer_python_dcu.txt diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt index 5271f78bb7..bf81d0baa8 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_mobile_v2.0_det +model_name:ch_ppocr_mobile_v2_0_det python:python3.7 gpu_list:192.168.0.1,192.168.0.2;0,1 Global.use_gpu:True diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt index 6b3352f741..df71e90702 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_mobile_v2.0_det +model_name:ch_ppocr_mobile_v2_0_det python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_mac_cpu_normal_normal_infer_python_mac_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_mac_cpu_normal_normal_infer_python_mac_cpu.txt similarity index 100% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_mac_cpu_normal_normal_infer_python_mac_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_mac_cpu_normal_normal_infer_python_mac_cpu.txt diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_pact_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_pact_infer_python.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_pact_infer_python.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_pact_infer_python.txt index 04c8d0e194..ba880d1f9e 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_pact_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_pact_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_mobile_v2.0_det_PACT +model_name:ch_ppocr_mobile_v2_0_det_PACT python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_ptq_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_ptq_infer_python.txt similarity index 93% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_ptq_infer_python.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_ptq_infer_python.txt index 2bdec84883..45c4fd1ae8 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_ptq_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_ptq_infer_python.txt @@ -1,5 +1,5 @@ ===========================kl_quant_params=========================== -model_name:ch_ppocr_mobile_v2.0_det_KL +model_name:ch_ppocr_mobile_v2_0_det_KL python:python3.7 Global.pretrained_model:null Global.save_inference_dir:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt similarity index 100% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det/train_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_FPGM/train_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_FPGM/train_infer_python.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det_FPGM/train_infer_python.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det_FPGM/train_infer_python.txt index dae3f8053a..0f6df1ac52 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_FPGM/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_FPGM/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_mobile_v2.0_det_FPGM +model_name:ch_ppocr_mobile_v2_0_det_FPGM python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt index 150a8a0315..2014c6dbcd 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_mobile_v2.0_det_FPGM +model_name:ch_ppocr_mobile_v2_0_det_FPGM python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt similarity index 92% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt index eb2fd0a001..f0e58dd566 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================cpp_infer_params=========================== -model_name:ch_ppocr_mobile_v2.0_det_KL +model_name:ch_ppocr_mobile_v2_0_det_KL use_opencv:True infer_model:./inference/ch_ppocr_mobile_v2.0_det_klquant_infer infer_quant:False diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_mac_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_mac_cpu.txt similarity index 89% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_mac_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_mac_cpu.txt index ea74b7fe76..2d8a680b72 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_mac_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_mac_cpu.txt @@ -1,5 +1,5 @@ ===========================kl_quant_params=========================== -infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ +infer_model:./inference/ch_ppocr_mobile_v2_0_det_infer/ infer_export:deploy/slim/quantization/quant_kl.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o infer_quant:True inference:tools/infer/predict_det.py diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_windows_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_windows_gpu_cpu.txt similarity index 89% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_windows_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_windows_gpu_cpu.txt index ea74b7fe76..2d8a680b72 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_windows_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_windows_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================kl_quant_params=========================== -infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ +infer_model:./inference/ch_ppocr_mobile_v2_0_det_infer/ infer_export:deploy/slim/quantization/quant_kl.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o infer_quant:True inference:tools/infer/predict_det.py diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt similarity index 95% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt index ab518de55a..c5dc52583b 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec_KL +model_name:ch_ppocr_mobile_v2_0_det_KL python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:./inference/ch_ppocr_mobile_v2.0_det_klquant_infer/ diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt similarity index 96% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt index 049ec78458..82d4db32ab 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_mobile_v2.0_det_KL +model_name:ch_ppocr_mobile_v2_0_det_KL python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:./inference/ch_ppocr_mobile_v2.0_det_klquant_infer/ diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_PACT/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_PACT/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt similarity index 92% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det_PACT/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det_PACT/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt index 17723f41ab..5132330597 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_PACT/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_PACT/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================cpp_infer_params=========================== -model_name:ch_ppocr_mobile_v2.0_det_PACT +model_name:ch_ppocr_mobile_v2_0_det_PACT use_opencv:True infer_model:./inference/ch_ppocr_mobile_v2.0_det_pact_infer infer_quant:False diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_PACT/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_PACT/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt similarity index 95% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det_PACT/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det_PACT/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt index 1a49a10f9b..3be53952f2 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_PACT/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_PACT/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_mobile_v2.0_det_PACT +model_name:ch_ppocr_mobile_v2_0_det_PACT python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:./inference/ch_ppocr_mobile_v2.0_det_pact_infer/ diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_PACT/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_PACT/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt similarity index 95% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det_PACT/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_det_PACT/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt index 909d738919..63e7f8f735 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_PACT/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_PACT/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_mobile_v2.0_det_PACT +model_name:ch_ppocr_mobile_v2_0_det_PACT python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:./inference/ch_ppocr_mobile_v2.0_det_pact_infer/ diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt similarity index 93% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt index 480fb16cdd..332e632bd5 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================cpp_infer_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec +model_name:ch_ppocr_mobile_v2_0_rec use_opencv:True infer_model:./inference/ch_ppocr_mobile_v2.0_rec_infer/ infer_quant:False diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt similarity index 93% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt index 5bab0c9e4c..78b76edae1 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt @@ -1,5 +1,5 @@ ===========================paddle2onnx_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec +model_name:ch_ppocr_mobile_v2_0_rec python:python3.7 2onnx: paddle2onnx --det_model_dir: diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt similarity index 96% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt index c0c5291cc4..5c60903f67 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec +model_name:ch_ppocr_mobile_v2_0_rec python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_infer_python.txt similarity index 98% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_infer_python.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_infer_python.txt index 36fdb1b91e..40f3979489 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec +model_name:ch_ppocr_mobile_v2_0_rec python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt index 631118c0a9..2f919d102b 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec +model_name:ch_ppocr_mobile_v2_0_rec python:python3.7 gpu_list:192.168.0.1,192.168.0.2;0,1 Global.use_gpu:True diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt index bd9c4a8df2..f60e2790e5 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec +model_name:ch_ppocr_mobile_v2_0_rec python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_pact_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_pact_infer_python.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_pact_infer_python.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_pact_infer_python.txt index 77472fbdfb..631518ffc4 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_pact_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_pact_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec_PACT +model_name:ch_ppocr_mobile_v2_0_rec_PACT python:python3.7 gpu_list:0 Global.use_gpu:True|True diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_ptq_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_ptq_infer_python.txt similarity index 94% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_ptq_infer_python.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_ptq_infer_python.txt index f63fe4c2bb..cd0828f3c6 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_ptq_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_ptq_infer_python.txt @@ -1,5 +1,5 @@ ===========================kl_quant_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec_KL +model_name:ch_ppocr_mobile_v2_0_rec_KL python:python3.7 Global.pretrained_model:null Global.save_inference_dir:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_FPGM/rec_chinese_lite_train_v2.0.yml b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/rec_chinese_lite_train_v2.0.yml similarity index 100% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec_FPGM/rec_chinese_lite_train_v2.0.yml rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/rec_chinese_lite_train_v2.0.yml diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_FPGM/train_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_infer_python.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec_FPGM/train_infer_python.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_infer_python.txt index 89daceeb5f..b2fb028823 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_FPGM/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec_FPGM +model_name:ch_ppocr_mobile_v2_0_rec_FPGM python:python3.7 gpu_list:0 Global.use_gpu:True|True diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt index 7abc3e9340..509240f6c5 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec_FPGM +model_name:ch_ppocr_mobile_v2_0_rec_FPGM python:python3.7 gpu_list:0 Global.use_gpu:True|True diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_KL/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt similarity index 93% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec_KL/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt index adf06257a7..ef4c93fcde 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_KL/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================cpp_infer_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec_KL +model_name:ch_ppocr_mobile_v2_0_rec_KL use_opencv:True infer_model:./inference/ch_ppocr_mobile_v2.0_rec_klquant_infer infer_quant:False diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_KL/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt similarity index 95% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec_KL/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt index d9de1cc19a..d904e22a7c 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_KL/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_mobile_v2.0_det_KL +model_name:ch_ppocr_mobile_v2_0_rec_KL python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:./inference/ch_ppocr_mobile_v2.0_det_klquant_infer/ diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_KL/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt similarity index 96% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec_KL/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt index 948e3dceb3..de4f7ed2c1 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_KL/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec_KL +model_name:ch_ppocr_mobile_v2_0_rec_KL python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/rec_chinese_lite_train_v2.0.yml b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_KL/rec_chinese_lite_train_v2.0.yml similarity index 100% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/rec_chinese_lite_train_v2.0.yml rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec_KL/rec_chinese_lite_train_v2.0.yml diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_PACT/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt similarity index 92% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec_PACT/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt index ba2df90f75..74ca7b50b8 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_PACT/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================cpp_infer_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec_PACT +model_name:ch_ppocr_mobile_v2_0_rec_PACT use_opencv:True infer_model:./inference/ch_ppocr_mobile_v2.0_rec_pact_infer infer_quant:False diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_PACT/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt similarity index 95% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec_PACT/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt index 229f70cf35..5a3047448b 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_PACT/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec_PACT +model_name:ch_ppocr_mobile_v2_0_rec_PACT python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:./inference/ch_ppocr_mobile_v2.0_det_pact_infer/ diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_PACT/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt similarity index 95% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec_PACT/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt index f123f36543..5871199bc0 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_PACT/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_mobile_v2.0_rec_PACT +model_name:ch_ppocr_mobile_v2_0_rec_PACT python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/rec_chinese_lite_train_v2.0.yml b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_PACT/rec_chinese_lite_train_v2.0.yml similarity index 100% rename from test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/rec_chinese_lite_train_v2.0.yml rename to test_tipc/configs/ch_ppocr_mobile_v2_0_rec_PACT/rec_chinese_lite_train_v2.0.yml diff --git a/test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt similarity index 94% rename from test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt index 7c980b2bae..ba8646fd9d 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================cpp_infer_params=========================== -model_name:ch_ppocr_server_v2.0 +model_name:ch_ppocr_server_v2_0 use_opencv:True infer_model:./inference/ch_ppocr_server_v2.0_det_infer/ infer_quant:False diff --git a/test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt similarity index 94% rename from test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt index b20596f7a1..53f8ab0e74 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================ch_ppocr_server_v2.0=========================== -model_name:ch_ppocr_server_v2.0 +model_name:ch_ppocr_server_v2_0 python:python3.7 infer_model:./inference/ch_ppocr_server_v2.0_det_infer/ infer_export:null diff --git a/test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt similarity index 95% rename from test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt index e478896a54..9e2cf191f3 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt @@ -1,5 +1,5 @@ ===========================paddle2onnx_params=========================== -model_name:ch_ppocr_server_v2.0 +model_name:ch_ppocr_server_v2_0 python:python3.7 2onnx: paddle2onnx --det_model_dir:./inference/ch_ppocr_server_v2.0_det_infer/ diff --git a/test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt similarity index 96% rename from test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt index bbfec44dba..55b27e04a2 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_serving_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_server_v2.0 +model_name:ch_ppocr_server_v2_0 python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:./inference/ch_ppocr_server_v2.0_det_infer/ diff --git a/test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt index 8853e709d4..21b8c9a082 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_server_v2.0 +model_name:ch_ppocr_server_v2_0 python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:./inference/ch_ppocr_server_v2.0_det_infer/ diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_det/det_r50_vd_db.yml b/test_tipc/configs/ch_ppocr_server_v2_0_det/det_r50_vd_db.yml similarity index 100% rename from test_tipc/configs/ch_ppocr_server_v2.0_det/det_r50_vd_db.yml rename to test_tipc/configs/ch_ppocr_server_v2_0_det/det_r50_vd_db.yml diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_det/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_det/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt similarity index 92% rename from test_tipc/configs/ch_ppocr_server_v2.0_det/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0_det/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt index 69ae939e2b..4a30affd07 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0_det/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_det/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================cpp_infer_params=========================== -model_name:ch_ppocr_server_v2.0_det +model_name:ch_ppocr_server_v2_0_det use_opencv:True infer_model:./inference/ch_ppocr_server_v2.0_det_infer/ infer_quant:False diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_det/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_det/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt similarity index 93% rename from test_tipc/configs/ch_ppocr_server_v2.0_det/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0_det/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt index c8bebf54f2..b7dd6e22b9 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0_det/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_det/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt @@ -1,5 +1,5 @@ ===========================paddle2onnx_params=========================== -model_name:ch_ppocr_server_v2.0_det +model_name:ch_ppocr_server_v2_0_det python:python3.7 2onnx: paddle2onnx --det_model_dir:./inference/ch_ppocr_server_v2.0_det_infer/ diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_det/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_det/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt similarity index 96% rename from test_tipc/configs/ch_ppocr_server_v2.0_det/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0_det/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt index 018dd1a227..4d4f0679bf 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0_det/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_det/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_server_v2.0_det +model_name:ch_ppocr_server_v2_0_det python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:./inference/ch_ppocr_server_v2.0_det_infer/ diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_det/train_infer_python.txt b/test_tipc/configs/ch_ppocr_server_v2_0_det/train_infer_python.txt similarity index 98% rename from test_tipc/configs/ch_ppocr_server_v2.0_det/train_infer_python.txt rename to test_tipc/configs/ch_ppocr_server_v2_0_det/train_infer_python.txt index 7b90a4078a..87e18ef7f3 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0_det/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_det/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_server_v2.0_det +model_name:ch_ppocr_server_v2_0_det python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_server_v2.0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt index 12388d9677..d1c7cf1d9e 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_server_v2.0_det +model_name:ch_ppocr_server_v2_0_det python:python3.7 gpu_list:192.168.0.1,192.168.0.2;0,1 Global.use_gpu:True diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_server_v2.0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt index 93ed14cb60..2802406b87 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_server_v2.0_det +model_name:ch_ppocr_server_v2_0_det python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_rec/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_rec/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt similarity index 93% rename from test_tipc/configs/ch_ppocr_server_v2.0_rec/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0_rec/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt index cbec272cce..3f3905516a 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0_rec/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_rec/model_linux_gpu_normal_normal_infer_cpp_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================cpp_infer_params=========================== -model_name:ch_ppocr_server_v2.0_rec +model_name:ch_ppocr_server_v2_0_rec use_opencv:True infer_model:./inference/ch_ppocr_server_v2.0_rec_infer/ infer_quant:False diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_rec/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_rec/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt similarity index 93% rename from test_tipc/configs/ch_ppocr_server_v2.0_rec/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0_rec/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt index 462f6090d9..89b9661003 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0_rec/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_rec/model_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt @@ -1,5 +1,5 @@ ===========================paddle2onnx_params=========================== -model_name:ch_ppocr_server_v2.0_rec +model_name:ch_ppocr_server_v2_0_rec python:python3.7 2onnx: paddle2onnx --det_model_dir: diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_rec/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_rec/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt similarity index 96% rename from test_tipc/configs/ch_ppocr_server_v2.0_rec/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0_rec/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt index 7f456320b6..4133e961cd 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0_rec/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_rec/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================serving_params=========================== -model_name:ch_ppocr_server_v2.0_rec +model_name:ch_ppocr_server_v2_0_rec python:python3.7 trans_model:-m paddle_serving_client.convert --det_dirname:null diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml b/test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml similarity index 100% rename from test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml rename to test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_rec/train_infer_python.txt b/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_infer_python.txt similarity index 98% rename from test_tipc/configs/ch_ppocr_server_v2.0_rec/train_infer_python.txt rename to test_tipc/configs/ch_ppocr_server_v2_0_rec/train_infer_python.txt index 9fc117d67c..e7cde1334a 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0_rec/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_server_v2.0_rec +model_name:ch_ppocr_server_v2_0_rec python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_server_v2.0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt index 9884ab247b..99b864d1b0 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_server_v2.0_rec +model_name:ch_ppocr_server_v2_0_rec python:python3.7 gpu_list:192.168.0.1,192.168.0.2;0,1 Global.use_gpu:True diff --git a/test_tipc/configs/ch_ppocr_server_v2.0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt similarity index 97% rename from test_tipc/configs/ch_ppocr_server_v2.0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt rename to test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt index 63ddaa4a8b..63aa07dd18 100644 --- a/test_tipc/configs/ch_ppocr_server_v2.0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:ch_ppocr_server_v2.0_rec +model_name:ch_ppocr_server_v2_0_rec python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/det_mv3_east_v2.0/det_mv3_east.yml b/test_tipc/configs/det_mv3_east_v2_0/det_mv3_east.yml similarity index 100% rename from test_tipc/configs/det_mv3_east_v2.0/det_mv3_east.yml rename to test_tipc/configs/det_mv3_east_v2_0/det_mv3_east.yml diff --git a/test_tipc/configs/det_mv3_east_v2.0/train_infer_python.txt b/test_tipc/configs/det_mv3_east_v2_0/train_infer_python.txt similarity index 98% rename from test_tipc/configs/det_mv3_east_v2.0/train_infer_python.txt rename to test_tipc/configs/det_mv3_east_v2_0/train_infer_python.txt index 1ec1597a4d..2818b5041c 100644 --- a/test_tipc/configs/det_mv3_east_v2.0/train_infer_python.txt +++ b/test_tipc/configs/det_mv3_east_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:det_mv3_east_v2.0 +model_name:det_mv3_east_v2_0 python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/det_mv3_pse_v2.0/det_mv3_pse.yml b/test_tipc/configs/det_mv3_pse_v2_0/det_mv3_pse.yml similarity index 100% rename from test_tipc/configs/det_mv3_pse_v2.0/det_mv3_pse.yml rename to test_tipc/configs/det_mv3_pse_v2_0/det_mv3_pse.yml diff --git a/test_tipc/configs/det_mv3_pse_v2.0/train_infer_python.txt b/test_tipc/configs/det_mv3_pse_v2_0/train_infer_python.txt similarity index 98% rename from test_tipc/configs/det_mv3_pse_v2.0/train_infer_python.txt rename to test_tipc/configs/det_mv3_pse_v2_0/train_infer_python.txt index daeec69f84..1fb4724ea4 100644 --- a/test_tipc/configs/det_mv3_pse_v2.0/train_infer_python.txt +++ b/test_tipc/configs/det_mv3_pse_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:det_mv3_pse_v2.0 +model_name:det_mv3_pse_v2_0 python:python3.7 gpu_list:0 Global.use_gpu:True|True diff --git a/test_tipc/configs/det_r50_db_v2.0/train_infer_python.txt b/test_tipc/configs/det_r50_db_v2_0/train_infer_python.txt similarity index 98% rename from test_tipc/configs/det_r50_db_v2.0/train_infer_python.txt rename to test_tipc/configs/det_r50_db_v2_0/train_infer_python.txt index 11af0ad18e..1d0d9693a9 100644 --- a/test_tipc/configs/det_r50_db_v2.0/train_infer_python.txt +++ b/test_tipc/configs/det_r50_db_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:det_r50_db_v2.0 +model_name:det_r50_db_v2_0 python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/det_r50_dcn_fce_ctw_v2.0/det_r50_vd_dcn_fce_ctw.yml b/test_tipc/configs/det_r50_dcn_fce_ctw_v2_0/det_r50_vd_dcn_fce_ctw.yml similarity index 100% rename from test_tipc/configs/det_r50_dcn_fce_ctw_v2.0/det_r50_vd_dcn_fce_ctw.yml rename to test_tipc/configs/det_r50_dcn_fce_ctw_v2_0/det_r50_vd_dcn_fce_ctw.yml diff --git a/test_tipc/configs/det_r50_dcn_fce_ctw_v2.0/train_infer_python.txt b/test_tipc/configs/det_r50_dcn_fce_ctw_v2_0/train_infer_python.txt similarity index 98% rename from test_tipc/configs/det_r50_dcn_fce_ctw_v2.0/train_infer_python.txt rename to test_tipc/configs/det_r50_dcn_fce_ctw_v2_0/train_infer_python.txt index 2d294fd303..4454d018ac 100644 --- a/test_tipc/configs/det_r50_dcn_fce_ctw_v2.0/train_infer_python.txt +++ b/test_tipc/configs/det_r50_dcn_fce_ctw_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:det_r50_dcn_fce_ctw_v2.0 +model_name:det_r50_dcn_fce_ctw_v2_0 python:python3.7 gpu_list:0 Global.use_gpu:True|True diff --git a/test_tipc/configs/det_r50_vd_sast_icdar15_v2.0/det_r50_vd_sast_icdar2015.yml b/test_tipc/configs/det_r50_vd_sast_icdar15_v2_0/det_r50_vd_sast_icdar2015.yml similarity index 100% rename from test_tipc/configs/det_r50_vd_sast_icdar15_v2.0/det_r50_vd_sast_icdar2015.yml rename to test_tipc/configs/det_r50_vd_sast_icdar15_v2_0/det_r50_vd_sast_icdar2015.yml diff --git a/test_tipc/configs/det_r50_vd_sast_icdar15_v2.0/train_infer_python.txt b/test_tipc/configs/det_r50_vd_sast_icdar15_v2_0/train_infer_python.txt similarity index 97% rename from test_tipc/configs/det_r50_vd_sast_icdar15_v2.0/train_infer_python.txt rename to test_tipc/configs/det_r50_vd_sast_icdar15_v2_0/train_infer_python.txt index b70ef46b4a..718aff271d 100644 --- a/test_tipc/configs/det_r50_vd_sast_icdar15_v2.0/train_infer_python.txt +++ b/test_tipc/configs/det_r50_vd_sast_icdar15_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:det_r50_vd_sast_icdar15_v2.0 +model_name:det_r50_vd_sast_icdar15_v2_0 python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/det_r50_vd_sast_totaltext_v2.0/det_r50_vd_sast_totaltext.yml b/test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/det_r50_vd_sast_totaltext.yml similarity index 100% rename from test_tipc/configs/det_r50_vd_sast_totaltext_v2.0/det_r50_vd_sast_totaltext.yml rename to test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/det_r50_vd_sast_totaltext.yml diff --git a/test_tipc/configs/det_r50_vd_sast_totaltext_v2.0/train_infer_python.txt b/test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/train_infer_python.txt similarity index 97% rename from test_tipc/configs/det_r50_vd_sast_totaltext_v2.0/train_infer_python.txt rename to test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/train_infer_python.txt index 7be5af7dde..cf176b3cff 100644 --- a/test_tipc/configs/det_r50_vd_sast_totaltext_v2.0/train_infer_python.txt +++ b/test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:det_r50_vd_sast_totaltext_v2.0 +model_name:det_r50_vd_sast_totaltext_v2_0 python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/rec_mv3_none_bilstm_ctc_v2.0/rec_icdar15_train.yml b/test_tipc/configs/rec_mv3_none_bilstm_ctc_v2_0/rec_icdar15_train.yml similarity index 100% rename from test_tipc/configs/rec_mv3_none_bilstm_ctc_v2.0/rec_icdar15_train.yml rename to test_tipc/configs/rec_mv3_none_bilstm_ctc_v2_0/rec_icdar15_train.yml diff --git a/test_tipc/configs/rec_mv3_none_bilstm_ctc_v2.0/train_infer_python.txt b/test_tipc/configs/rec_mv3_none_bilstm_ctc_v2_0/train_infer_python.txt similarity index 98% rename from test_tipc/configs/rec_mv3_none_bilstm_ctc_v2.0/train_infer_python.txt rename to test_tipc/configs/rec_mv3_none_bilstm_ctc_v2_0/train_infer_python.txt index 4e34a6a525..9e1fb14a0a 100644 --- a/test_tipc/configs/rec_mv3_none_bilstm_ctc_v2.0/train_infer_python.txt +++ b/test_tipc/configs/rec_mv3_none_bilstm_ctc_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:rec_mv3_none_bilstm_ctc_v2.0 +model_name:rec_mv3_none_bilstm_ctc_v2_0 python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/rec_mv3_none_none_ctc_v2.0/rec_icdar15_train.yml b/test_tipc/configs/rec_mv3_none_none_ctc_v2_0/rec_icdar15_train.yml similarity index 100% rename from test_tipc/configs/rec_mv3_none_none_ctc_v2.0/rec_icdar15_train.yml rename to test_tipc/configs/rec_mv3_none_none_ctc_v2_0/rec_icdar15_train.yml diff --git a/test_tipc/configs/rec_mv3_none_none_ctc_v2.0/train_infer_python.txt b/test_tipc/configs/rec_mv3_none_none_ctc_v2_0/train_infer_python.txt similarity index 97% rename from test_tipc/configs/rec_mv3_none_none_ctc_v2.0/train_infer_python.txt rename to test_tipc/configs/rec_mv3_none_none_ctc_v2_0/train_infer_python.txt index 593de3ff20..09fa408cab 100644 --- a/test_tipc/configs/rec_mv3_none_none_ctc_v2.0/train_infer_python.txt +++ b/test_tipc/configs/rec_mv3_none_none_ctc_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:rec_mv3_none_none_ctc_v2.0 +model_name:rec_mv3_none_none_ctc_v2_0 python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/rec_mv3_tps_bilstm_att_v2.0/rec_mv3_tps_bilstm_att.yml b/test_tipc/configs/rec_mv3_tps_bilstm_att_v2_0/rec_mv3_tps_bilstm_att.yml similarity index 100% rename from test_tipc/configs/rec_mv3_tps_bilstm_att_v2.0/rec_mv3_tps_bilstm_att.yml rename to test_tipc/configs/rec_mv3_tps_bilstm_att_v2_0/rec_mv3_tps_bilstm_att.yml diff --git a/test_tipc/configs/rec_mv3_tps_bilstm_att_v2.0/train_infer_python.txt b/test_tipc/configs/rec_mv3_tps_bilstm_att_v2_0/train_infer_python.txt similarity index 97% rename from test_tipc/configs/rec_mv3_tps_bilstm_att_v2.0/train_infer_python.txt rename to test_tipc/configs/rec_mv3_tps_bilstm_att_v2_0/train_infer_python.txt index 1b2d9abb0f..6a1d76810c 100644 --- a/test_tipc/configs/rec_mv3_tps_bilstm_att_v2.0/train_infer_python.txt +++ b/test_tipc/configs/rec_mv3_tps_bilstm_att_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:rec_mv3_tps_bilstm_att_v2.0 +model_name:rec_mv3_tps_bilstm_att_v2_0 python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2.0/rec_icdar15_train.yml b/test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2_0/rec_icdar15_train.yml similarity index 100% rename from test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2.0/rec_icdar15_train.yml rename to test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2_0/rec_icdar15_train.yml diff --git a/test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2.0/train_infer_python.txt b/test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2_0/train_infer_python.txt similarity index 97% rename from test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2.0/train_infer_python.txt rename to test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2_0/train_infer_python.txt index 1367c7abd4..1b6be1b04b 100644 --- a/test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2.0/train_infer_python.txt +++ b/test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:rec_mv3_tps_bilstm_ctc_v2.0 +model_name:rec_mv3_tps_bilstm_ctc_v2_0 python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/rec_r32_gaspin_bilstm_att/rec_r32_gaspin_bilstm_att.yml b/test_tipc/configs/rec_r32_gaspin_bilstm_att/rec_r32_gaspin_bilstm_att.yml index d0cb20481f..21d56b685c 100644 --- a/test_tipc/configs/rec_r32_gaspin_bilstm_att/rec_r32_gaspin_bilstm_att.yml +++ b/test_tipc/configs/rec_r32_gaspin_bilstm_att/rec_r32_gaspin_bilstm_att.yml @@ -8,7 +8,7 @@ Global: # evaluation is run every 5000 iterations after the 4000th iteration eval_batch_step: [0, 2000] cal_metric_during_train: True - pretrained_model: + pretrained_model: pretrain_models/rec_r32_gaspin_bilstm_att_train/best_accuracy checkpoints: save_inference_dir: use_visualdl: False diff --git a/test_tipc/configs/rec_r32_gaspin_bilstm_att/train_infer_python.txt b/test_tipc/configs/rec_r32_gaspin_bilstm_att/train_infer_python.txt index 4915055a57..115dfd661a 100644 --- a/test_tipc/configs/rec_r32_gaspin_bilstm_att/train_infer_python.txt +++ b/test_tipc/configs/rec_r32_gaspin_bilstm_att/train_infer_python.txt @@ -1,6 +1,6 @@ ===========================train_params=========================== model_name:rec_r32_gaspin_bilstm_att -python:python +python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True Global.auto_cast:null @@ -39,11 +39,11 @@ infer_export:tools/export_model.py -c test_tipc/configs/rec_r32_gaspin_bilstm_at infer_quant:False inference:tools/infer/predict_rec.py --rec_char_dict_path=./ppocr/utils/dict/spin_dict.txt --use_space_char=False --rec_image_shape="3,32,100" --rec_algorithm="SPIN" --use_gpu:True|False ---enable_mkldnn:True|False ---cpu_threads:1|6 +--enable_mkldnn:False +--cpu_threads:6 --rec_batch_num:1|6 ---use_tensorrt:False|False ---precision:fp32|int8 +--use_tensorrt:False +--precision:fp32 --rec_model_dir: --image_dir:./inference/rec_inference --save_log_path:./test/output/ diff --git a/test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2.0/rec_icdar15_train.yml b/test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2_0/rec_icdar15_train.yml similarity index 100% rename from test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2.0/rec_icdar15_train.yml rename to test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2_0/rec_icdar15_train.yml diff --git a/test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2.0/train_infer_python.txt b/test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2_0/train_infer_python.txt similarity index 97% rename from test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2.0/train_infer_python.txt rename to test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2_0/train_infer_python.txt index 46aa3d7190..98953b31fa 100644 --- a/test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2.0/train_infer_python.txt +++ b/test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:rec_r34_vd_none_bilstm_ctc_v2.0 +model_name:rec_r34_vd_none_bilstm_ctc_v2_0 python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/rec_r34_vd_none_none_ctc_v2.0/rec_icdar15_train.yml b/test_tipc/configs/rec_r34_vd_none_none_ctc_v2_0/rec_icdar15_train.yml similarity index 100% rename from test_tipc/configs/rec_r34_vd_none_none_ctc_v2.0/rec_icdar15_train.yml rename to test_tipc/configs/rec_r34_vd_none_none_ctc_v2_0/rec_icdar15_train.yml diff --git a/test_tipc/configs/rec_r34_vd_none_none_ctc_v2.0/train_infer_python.txt b/test_tipc/configs/rec_r34_vd_none_none_ctc_v2_0/train_infer_python.txt similarity index 97% rename from test_tipc/configs/rec_r34_vd_none_none_ctc_v2.0/train_infer_python.txt rename to test_tipc/configs/rec_r34_vd_none_none_ctc_v2_0/train_infer_python.txt index 3e066d7b72..37e66e8266 100644 --- a/test_tipc/configs/rec_r34_vd_none_none_ctc_v2.0/train_infer_python.txt +++ b/test_tipc/configs/rec_r34_vd_none_none_ctc_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:rec_r34_vd_none_none_ctc_v2.0 +model_name:rec_r34_vd_none_none_ctc_v2_0 python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2.0/rec_r34_vd_tps_bilstm_att.yml b/test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2_0/rec_r34_vd_tps_bilstm_att.yml similarity index 100% rename from test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2.0/rec_r34_vd_tps_bilstm_att.yml rename to test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2_0/rec_r34_vd_tps_bilstm_att.yml diff --git a/test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2.0/train_infer_python.txt b/test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2_0/train_infer_python.txt similarity index 97% rename from test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2.0/train_infer_python.txt rename to test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2_0/train_infer_python.txt index 1e4f46633e..772db83770 100644 --- a/test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2.0/train_infer_python.txt +++ b/test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:rec_r34_vd_tps_bilstm_att_v2.0 +model_name:rec_r34_vd_tps_bilstm_att_v2_0 python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2.0/rec_icdar15_train.yml b/test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2_0/rec_icdar15_train.yml similarity index 100% rename from test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2.0/rec_icdar15_train.yml rename to test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2_0/rec_icdar15_train.yml diff --git a/test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2.0/train_infer_python.txt b/test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2_0/train_infer_python.txt similarity index 97% rename from test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2.0/train_infer_python.txt rename to test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2_0/train_infer_python.txt index 9e795b6645..bc1ea7e851 100644 --- a/test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2.0/train_infer_python.txt +++ b/test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2_0/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:rec_r34_vd_tps_bilstm_ctc_v2.0 +model_name:rec_r34_vd_tps_bilstm_ctc_v2_0 python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/docs/benchmark_train.md b/test_tipc/docs/benchmark_train.md index a7f95eb6c5..0196edc247 100644 --- a/test_tipc/docs/benchmark_train.md +++ b/test_tipc/docs/benchmark_train.md @@ -72,4 +72,4 @@ train_log/ | PP-Structure-table |[config](../configs/en_table_structure/train_infer_python.txt) | 14.151 | 14.077 / 14.23 / 14.25 |0.012140351 | 16.285 | 16.595 / 16.878 / 16.531 | 0.020559308 | 20,000| 5,000| | det_r50_dcn_fce_ctw_v2.0 |[config](../configs/det_r50_dcn_fce_ctw_v2.0/train_infer_python.txt) | 14.057 | 14.029 / 14.02 / 14.014 |0.001069214 | 18.298 |18.411 / 18.376 / 18.331| 0.004345228 | 10,000| 2,000| | ch_PP-OCRv3_det |[config](../configs/ch_PP-OCRv3_det/train_infer_python.txt) | 8.622 | 8.431 / 8.423 / 8.479|0.006604552 | 14.203 |14.346 14.468 14.23| 0.016450097 | 10,000| 2,000| -| ch_PP-OCRv3_rec |[config](../configs/ch_PP-OCRv3_rec/train_infer_python.txt) | 73.627 | 72.46 / 73.575 / 73.704|0.016878324 | | | | 160,000| 40,000| \ No newline at end of file +| ch_PP-OCRv3_rec |[config](../configs/ch_PP-OCRv3_rec/train_infer_python.txt) | 90.239 | 90.077 / 91.513 / 91.325|0.01569176 | | | | 160,000| 40,000| diff --git a/test_tipc/prepare.sh b/test_tipc/prepare.sh index cb3fa2440d..88154d6187 100644 --- a/test_tipc/prepare.sh +++ b/test_tipc/prepare.sh @@ -22,7 +22,7 @@ trainer_list=$(func_parser_value "${lines[14]}") if [ ${MODE} = "benchmark_train" ];then pip install -r requirements.txt - if [[ ${model_name} =~ "ch_ppocr_mobile_v2.0_det" || ${model_name} =~ "det_mv3_db_v2_0" ]];then + if [[ ${model_name} =~ "ch_ppocr_mobile_v2_0_det" || ${model_name} =~ "det_mv3_db_v2_0" ]];then wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/pretrained/MobileNetV3_large_x0_5_pretrained.pdparams --no-check-certificate rm -rf ./train_data/icdar2015 wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dataset/icdar2015_benckmark.tar --no-check-certificate @@ -30,7 +30,7 @@ if [ ${MODE} = "benchmark_train" ];then ln -s ./icdar2015_benckmark ./icdar2015 cd ../ fi - if [[ ${model_name} =~ "ch_ppocr_server_v2.0_det" || ${model_name} =~ "ch_PP-OCRv3_det" ]];then + if [[ ${model_name} =~ "ch_ppocr_server_v2_0_det" || ${model_name} =~ "ch_PP-OCRv3_det" ]];then rm -rf ./train_data/icdar2015 wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dataset/icdar2015_benckmark.tar --no-check-certificate cd ./train_data/ && tar xf icdar2015_benckmark.tar @@ -55,7 +55,7 @@ if [ ${MODE} = "benchmark_train" ];then ln -s ./icdar2015_benckmark ./icdar2015 cd ../ fi - if [[ ${model_name} =~ "det_r50_db_v2.0" || ${model_name} =~ "det_r50_vd_pse_v2_0" ]];then + if [[ ${model_name} =~ "det_r50_db_v2_0" || ${model_name} =~ "det_r50_vd_pse_v2_0" ]];then wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/pretrained/ResNet50_vd_ssld_pretrained.pdparams --no-check-certificate rm -rf ./train_data/icdar2015 wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dataset/icdar2015_benckmark.tar --no-check-certificate @@ -71,13 +71,23 @@ if [ ${MODE} = "benchmark_train" ];then ln -s ./icdar2015_benckmark ./icdar2015 cd ../ fi - if [[ ${model_name} =~ "ch_ppocr_mobile_v2.0_rec" || ${model_name} =~ "ch_ppocr_server_v2.0_rec" || ${model_name} =~ "ch_PP-OCRv2_rec" || ${model_name} =~ "rec_mv3_none_bilstm_ctc_v2.0" || ${model_name} =~ "ch_PP-OCRv3_rec" ]];then - rm -rf ./train_data/ic15_data_benckmark + if [[ ${model_name} =~ "ch_ppocr_mobile_v2_0_rec" || ${model_name} =~ "ch_ppocr_server_v2_0_rec" || ${model_name} =~ "ch_PP-OCRv2_rec" || ${model_name} =~ "rec_mv3_none_bilstm_ctc_v2_0" || ${model_name} =~ "ch_PP-OCRv3_rec" ]];then + rm -rf ./train_data/ic15_data wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dataset/ic15_data_benckmark.tar --no-check-certificate cd ./train_data/ && tar xf ic15_data_benckmark.tar ln -s ./ic15_data_benckmark ./ic15_data cd ../ fi + if [[ ${model_name} =~ "ch_PP-OCRv2_rec" || ${model_name} =~ "ch_PP-OCRv3_rec" ]];then + rm -rf ./train_data/ic15_data + wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dataset/ic15_data_benckmark.tar --no-check-certificate + cd ./train_data/ && tar xf ic15_data_benckmark.tar + ln -s ./ic15_data_benckmark ./ic15_data + cd ic15_data + mv rec_gt_train4w.txt rec_gt_train.txt + cd ../ + cd ../ + fi if [[ ${model_name} == "en_table_structure" ]];then wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/dygraph_v2.1/table/en_ppocr_mobile_v2.0_table_structure_train.tar --no-check-certificate cd ./pretrain_models/ && tar xf en_ppocr_mobile_v2.0_table_structure_train.tar && cd ../ @@ -87,7 +97,7 @@ if [ ${MODE} = "benchmark_train" ];then ln -s ./pubtabnet_benckmark ./pubtabnet cd ../ fi - if [[ ${model_name} == "det_r50_dcn_fce_ctw_v2.0" ]]; then + if [[ ${model_name} == "det_r50_dcn_fce_ctw_v2_0" ]]; then wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/contribution/det_r50_dcn_fce_ctw_v2.0_train.tar --no-check-certificate cd ./pretrain_models/ && tar xf det_r50_dcn_fce_ctw_v2.0_train.tar && cd ../ rm -rf ./train_data/icdar2015 @@ -161,7 +171,7 @@ if [ ${MODE} = "lite_train_lite_infer" ];then cd ./pretrain_models/ && tar xf en_server_pgnetA.tar && cd ../ cd ./train_data && tar xf total_text_lite.tar && ln -s total_text_lite total_text && cd ../ fi - if [ ${model_name} == "det_r50_vd_sast_icdar15_v2.0" ] || [ ${model_name} == "det_r50_vd_sast_totaltext_v2.0" ]; then + if [ ${model_name} == "det_r50_vd_sast_icdar15_v2_0" ] || [ ${model_name} == "det_r50_vd_sast_totaltext_v2_0" ]; then wget -nc -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/ResNet50_vd_ssld_pretrained.pdparams --no-check-certificate wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_sast_icdar15_v2.0_train.tar --no-check-certificate wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/total_text_lite.tar --no-check-certificate @@ -172,16 +182,16 @@ if [ ${MODE} = "lite_train_lite_infer" ];then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_mv3_db_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf det_mv3_db_v2.0_train.tar && cd ../ fi - if [ ${model_name} == "det_r50_db_v2.0" ]; then + if [ ${model_name} == "det_r50_db_v2_0" ]; then wget -nc -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/ResNet50_vd_ssld_pretrained.pdparams --no-check-certificate wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_db_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf det_r50_vd_db_v2.0_train.tar && cd ../ fi - if [ ${model_name} == "ch_ppocr_mobile_v2.0_rec_FPGM" ]; then + if [ ${model_name} == "ch_ppocr_mobile_v2_0_rec_FPGM" ]; then wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_train.tar --no-check-certificate cd ./pretrain_models/ && tar xf ch_ppocr_mobile_v2.0_rec_train.tar && cd ../ fi - if [ ${model_name} == "det_mv3_east_v2.0" ]; then + if [ ${model_name} == "det_mv3_east_v2_0" ]; then wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_mv3_east_v2.0_train.tar --no-check-certificate cd ./pretrain_models/ && tar xf det_mv3_east_v2.0_train.tar && cd ../ fi @@ -189,10 +199,14 @@ if [ ${MODE} = "lite_train_lite_infer" ];then wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_east_v2.0_train.tar --no-check-certificate cd ./pretrain_models/ && tar xf det_r50_vd_east_v2.0_train.tar && cd ../ fi - if [ ${model_name} == "det_r50_dcn_fce_ctw_v2.0" ]; then + if [ ${model_name} == "det_r50_dcn_fce_ctw_v2_0" ]; then wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/contribution/det_r50_dcn_fce_ctw_v2.0_train.tar --no-check-certificate cd ./pretrain_models/ && tar xf det_r50_dcn_fce_ctw_v2.0_train.tar & cd ../ fi + if [ ${model_name} == "rec_r32_gaspin_bilstm_att" ]; then + wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/rec_r32_gaspin_bilstm_att_train.tar --no-check-certificate + cd ./pretrain_models/ && tar xf rec_r32_gaspin_bilstm_att_train.tar && cd ../ + fi elif [ ${MODE} = "whole_train_whole_infer" ];then wget -nc -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/MobileNetV3_large_x0_5_pretrained.pdparams --no-check-certificate @@ -220,7 +234,7 @@ elif [ ${MODE} = "whole_train_whole_infer" ];then cd ./pretrain_models/ && tar xf en_server_pgnetA.tar && cd ../ cd ./train_data && tar xf total_text.tar && ln -s total_text_lite total_text && cd ../ fi - if [ ${model_name} == "det_r50_vd_sast_totaltext_v2.0" ]; then + if [ ${model_name} == "det_r50_vd_sast_totaltext_v2_0" ]; then wget -nc -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/ResNet50_vd_ssld_pretrained.pdparams --no-check-certificate wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/total_text_lite.tar --no-check-certificate cd ./train_data && tar xf total_text.tar && ln -s total_text_lite total_text && cd ../ @@ -264,32 +278,32 @@ elif [ ${MODE} = "whole_infer" ];then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/rec_inference.tar --no-check-certificate cd ./inference && tar xf rec_inference.tar && tar xf ch_det_data_50.tar && cd ../ - if [ ${model_name} = "ch_ppocr_mobile_v2.0_det" ]; then + if [ ${model_name} = "ch_ppocr_mobile_v2_0_det" ]; then eval_model_name="ch_ppocr_mobile_v2.0_det_train" rm -rf ./train_data/icdar2015 wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar --no-check-certificate cd ./inference && tar xf ${eval_model_name}.tar && tar xf ch_det_data_50.tar && tar xf ch_ppocr_mobile_v2.0_det_infer.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_mobile_v2.0_det_PACT" ]; then + elif [ ${model_name} = "ch_ppocr_mobile_v2_0_det_PACT" ]; then eval_model_name="ch_ppocr_mobile_v2.0_det_prune_infer" wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/slim/ch_ppocr_mobile_v2.0_det_prune_infer.tar --no-check-certificate cd ./inference && tar xf ${eval_model_name}.tar && tar xf ch_det_data_50.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_server_v2.0_det" ]; then + elif [ ${model_name} = "ch_ppocr_server_v2_0_det" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_server_v2.0_det_train.tar && tar xf ch_det_data_50.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_mobile_v2.0" ]; then + elif [ ${model_name} = "ch_ppocr_mobile_v2_0" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_det_infer.tar && tar xf ch_ppocr_mobile_v2.0_rec_infer.tar && tar xf ch_det_data_50.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_server_v2.0" ]; then + elif [ ${model_name} = "ch_ppocr_server_v2_0" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_server_v2.0_det_infer.tar && tar xf ch_ppocr_server_v2.0_rec_infer.tar && tar xf ch_det_data_50.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_mobile_v2.0_rec_PACT" ]; then + elif [ ${model_name} = "ch_ppocr_mobile_v2_0_rec_PACT" ]; then eval_model_name="ch_ppocr_mobile_v2.0_rec_slim_infer" wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_infer.tar --no-check-certificate cd ./inference && tar xf ${eval_model_name}.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_mobile_v2.0_rec_FPGM" ]; then + elif [ ${model_name} = "ch_ppocr_mobile_v2_0_rec_FPGM" ]; then eval_model_name="ch_PP-OCRv2_rec_infer" wget -nc -P ./inference https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_infer.tar --no-check-certificate cd ./inference && tar xf ${eval_model_name}.tar && cd ../ @@ -334,39 +348,39 @@ elif [ ${MODE} = "whole_infer" ];then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/pgnet/en_server_pgnetA.tar --no-check-certificate cd ./inference && tar xf en_server_pgnetA.tar && tar xf ch_det_data_50.tar && cd ../ fi - if [ ${model_name} == "det_r50_vd_sast_icdar15_v2.0" ]; then + if [ ${model_name} == "det_r50_vd_sast_icdar15_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_sast_icdar15_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf det_r50_vd_sast_icdar15_v2.0_train.tar && tar xf ch_det_data_50.tar && cd ../ fi - if [ ${model_name} == "rec_mv3_none_none_ctc_v2.0" ]; then + if [ ${model_name} == "rec_mv3_none_none_ctc_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_none_none_ctc_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf rec_mv3_none_none_ctc_v2.0_train.tar && cd ../ fi - if [ ${model_name} == "rec_r34_vd_none_none_ctc_v2.0" ]; then + if [ ${model_name} == "rec_r34_vd_none_none_ctc_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_none_none_ctc_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf rec_r34_vd_none_none_ctc_v2.0_train.tar && cd ../ fi - if [ ${model_name} == "rec_mv3_none_bilstm_ctc_v2.0" ]; then + if [ ${model_name} == "rec_mv3_none_bilstm_ctc_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_none_bilstm_ctc_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf rec_mv3_none_bilstm_ctc_v2.0_train.tar && cd ../ fi - if [ ${model_name} == "rec_r34_vd_none_bilstm_ctc_v2.0" ]; then + if [ ${model_name} == "rec_r34_vd_none_bilstm_ctc_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_none_bilstm_ctc_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf rec_r34_vd_none_bilstm_ctc_v2.0_train.tar && cd ../ fi - if [ ${model_name} == "rec_mv3_tps_bilstm_ctc_v2.0" ]; then + if [ ${model_name} == "rec_mv3_tps_bilstm_ctc_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_tps_bilstm_ctc_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf rec_mv3_tps_bilstm_ctc_v2.0_train.tar && cd ../ fi - if [ ${model_name} == "rec_r34_vd_tps_bilstm_ctc_v2.0" ]; then + if [ ${model_name} == "rec_r34_vd_tps_bilstm_ctc_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_tps_bilstm_ctc_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf rec_r34_vd_tps_bilstm_ctc_v2.0_train.tar && cd ../ fi - if [ ${model_name} == "ch_ppocr_server_v2.0_rec" ]; then + if [ ${model_name} == "ch_ppocr_server_v2_0_rec" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_train.tar --no-check-certificate cd ./inference/ && tar xf ch_ppocr_server_v2.0_rec_train.tar && cd ../ fi - if [ ${model_name} == "ch_ppocr_mobile_v2.0_rec" ]; then + if [ ${model_name} == "ch_ppocr_mobile_v2_0_rec" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_train.tar --no-check-certificate cd ./inference/ && tar xf ch_ppocr_mobile_v2.0_rec_train.tar && cd ../ fi @@ -374,11 +388,11 @@ elif [ ${MODE} = "whole_infer" ];then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mtb_nrtr_train.tar --no-check-certificate cd ./inference/ && tar xf rec_mtb_nrtr_train.tar && cd ../ fi - if [ ${model_name} == "rec_mv3_tps_bilstm_att_v2.0" ]; then + if [ ${model_name} == "rec_mv3_tps_bilstm_att_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_tps_bilstm_att_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf rec_mv3_tps_bilstm_att_v2.0_train.tar && cd ../ fi - if [ ${model_name} == "rec_r34_vd_tps_bilstm_att_v2.0" ]; then + if [ ${model_name} == "rec_r34_vd_tps_bilstm_att_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_tps_bilstm_att_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf rec_r34_vd_tps_bilstm_att_v2.0_train.tar && cd ../ fi @@ -391,7 +405,7 @@ elif [ ${MODE} = "whole_infer" ];then cd ./inference/ && tar xf rec_r50_vd_srn_train.tar && cd ../ fi - if [ ${model_name} == "det_r50_vd_sast_totaltext_v2.0" ]; then + if [ ${model_name} == "det_r50_vd_sast_totaltext_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_sast_totaltext_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf det_r50_vd_sast_totaltext_v2.0_train.tar && cd ../ fi @@ -399,11 +413,11 @@ elif [ ${MODE} = "whole_infer" ];then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_mv3_db_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf det_mv3_db_v2.0_train.tar && tar xf ch_det_data_50.tar && cd ../ fi - if [ ${model_name} == "det_r50_db_v2.0" ]; then + if [ ${model_name} == "det_r50_db_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_db_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf det_r50_vd_db_v2.0_train.tar && tar xf ch_det_data_50.tar && cd ../ fi - if [ ${model_name} == "det_mv3_pse_v2.0" ]; then + if [ ${model_name} == "det_mv3_pse_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.1/en_det/det_mv3_pse_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf det_mv3_pse_v2.0_train.tar & cd ../ fi @@ -411,7 +425,7 @@ elif [ ${MODE} = "whole_infer" ];then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.1/en_det/det_r50_vd_pse_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf det_r50_vd_pse_v2.0_train.tar & cd ../ fi - if [ ${model_name} == "det_mv3_east_v2.0" ]; then + if [ ${model_name} == "det_mv3_east_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_mv3_east_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf det_mv3_east_v2.0_train.tar & cd ../ fi @@ -419,7 +433,7 @@ elif [ ${MODE} = "whole_infer" ];then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_east_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf det_r50_vd_east_v2.0_train.tar & cd ../ fi - if [ ${model_name} == "det_r50_dcn_fce_ctw_v2.0" ]; then + if [ ${model_name} == "det_r50_dcn_fce_ctw_v2_0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/contribution/det_r50_dcn_fce_ctw_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf det_r50_dcn_fce_ctw_v2.0_train.tar & cd ../ fi @@ -434,7 +448,7 @@ fi if [[ ${model_name} =~ "KL" ]]; then wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/icdar2015_lite.tar --no-check-certificate cd ./train_data/ && tar xf icdar2015_lite.tar && rm -rf ./icdar2015 && ln -s ./icdar2015_lite ./icdar2015 && cd ../ - if [ ${model_name} = "ch_ppocr_mobile_v2.0_det_KL" ]; then + if [ ${model_name} = "ch_ppocr_mobile_v2_0_det_KL" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_det_infer.tar && tar xf ch_det_data_50.tar && cd ../ @@ -466,7 +480,7 @@ if [[ ${model_name} =~ "KL" ]]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_infer.tar --no-check-certificate cd ./inference && tar xf ch_PP-OCRv3_det_infer.tar && tar xf ch_det_data_50.tar && cd ../ fi - if [ ${model_name} = "ch_ppocr_mobile_v2.0_rec_KL" ]; then + if [ ${model_name} = "ch_ppocr_mobile_v2_0_rec_KL" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar --no-check-certificate wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/rec_inference.tar --no-check-certificate wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ic15_data.tar --no-check-certificate @@ -484,35 +498,35 @@ if [[ ${model_name} =~ "KL" ]]; then fi if [ ${MODE} = "cpp_infer" ];then - if [ ${model_name} = "ch_ppocr_mobile_v2.0_det" ]; then + if [ ${model_name} = "ch_ppocr_mobile_v2_0_det" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_det_infer.tar && tar xf ch_det_data_50.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_mobile_v2.0_det_KL" ]; then + elif [ ${model_name} = "ch_ppocr_mobile_v2_0_det_KL" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/tipc_fake_model/ch_ppocr_mobile_v2.0_det_klquant_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_det_klquant_infer.tar && tar xf ch_det_data_50.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_mobile_v2.0_det_PACT" ]; then + elif [ ${model_name} = "ch_ppocr_mobile_v2_0_det_PACT" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/tipc_fake_model/ch_ppocr_mobile_v2.0_det_pact_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_det_pact_infer.tar && tar xf ch_det_data_50.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_mobile_v2.0_rec" ]; then + elif [ ${model_name} = "ch_ppocr_mobile_v2_0_rec" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/rec_inference.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_rec_infer.tar && tar xf rec_inference.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_mobile_v2.0_rec_KL" ]; then + elif [ ${model_name} = "ch_ppocr_mobile_v2_0_rec_KL" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/rec_inference.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/tipc_fake_model/ch_ppocr_mobile_v2.0_rec_klquant_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_rec_klquant_infer.tar && tar xf rec_inference.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_mobile_v2.0_rec_PACT" ]; then + elif [ ${model_name} = "ch_ppocr_mobile_v2_0_rec_PACT" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/rec_inference.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/tipc_fake_model/ch_ppocr_mobile_v2.0_rec_pact_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_rec_pact_infer.tar && tar xf rec_inference.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_server_v2.0_det" ]; then + elif [ ${model_name} = "ch_ppocr_server_v2_0_det" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_server_v2.0_det_infer.tar && tar xf ch_det_data_50.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_server_v2.0_rec" ]; then + elif [ ${model_name} = "ch_ppocr_server_v2_0_rec" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/rec_inference.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_server_v2.0_rec_infer.tar && tar xf rec_inference.tar && cd ../ @@ -564,12 +578,12 @@ if [ ${MODE} = "cpp_infer" ];then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/rec_inference.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/tipc_fake_model/ch_PP-OCRv3_rec_pact_infer.tar --no-check-certificate cd ./inference && tar xf ch_PP-OCRv3_rec_pact_infer.tar && tar xf rec_inference.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_mobile_v2.0" ]; then + elif [ ${model_name} = "ch_ppocr_mobile_v2_0" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_det_infer.tar && tar xf ch_ppocr_mobile_v2.0_rec_infer.tar && tar xf ch_det_data_50.tar && cd ../ - elif [ ${model_name} = "ch_ppocr_server_v2.0" ]; then + elif [ ${model_name} = "ch_ppocr_server_v2_0" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar --no-check-certificate @@ -597,7 +611,7 @@ if [ ${MODE} = "serving_infer" ];then ${python_name} -m pip install paddle_serving_client ${python_name} -m pip install paddle-serving-app # wget model - if [ ${model_name} == "ch_ppocr_mobile_v2.0_det_KL" ] || [ ${model_name} == "ch_ppocr_mobile_v2.0_rec_KL" ] ; then + if [ ${model_name} == "ch_ppocr_mobile_v2_0_det_KL" ] || [ ${model_name} == "ch_ppocr_mobile_v2.0_rec_KL" ] ; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/tipc_fake_model/ch_ppocr_mobile_v2.0_det_klquant_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/tipc_fake_model/ch_ppocr_mobile_v2.0_rec_klquant_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_det_klquant_infer.tar && tar xf ch_ppocr_mobile_v2.0_rec_klquant_infer.tar && cd ../ @@ -609,7 +623,7 @@ if [ ${MODE} = "serving_infer" ];then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/tipc_fake_model/ch_PP-OCRv3_det_klquant_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/tipc_fake_model/ch_PP-OCRv3_rec_klquant_infer.tar --no-check-certificate cd ./inference && tar xf ch_PP-OCRv3_det_klquant_infer.tar && tar xf ch_PP-OCRv3_rec_klquant_infer.tar && cd ../ - elif [ ${model_name} == "ch_ppocr_mobile_v2.0_det_PACT" ] || [ ${model_name} == "ch_ppocr_mobile_v2.0_rec_PACT" ] ; then + elif [ ${model_name} == "ch_ppocr_mobile_v2_0_det_PACT" ] || [ ${model_name} == "ch_ppocr_mobile_v2.0_rec_PACT" ] ; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/tipc_fake_model/ch_ppocr_mobile_v2.0_det_pact_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/tipc_fake_model/ch_ppocr_mobile_v2.0_rec_pact_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_det_pact_infer.tar && tar xf ch_ppocr_mobile_v2.0_rec_pact_infer.tar && cd ../ @@ -621,11 +635,11 @@ if [ ${MODE} = "serving_infer" ];then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/tipc_fake_model/ch_PP-OCRv3_det_pact_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/tipc_fake_model/ch_PP-OCRv3_rec_pact_infer.tar --no-check-certificate cd ./inference && tar xf ch_PP-OCRv3_det_pact_infer.tar && tar xf ch_PP-OCRv3_rec_pact_infer.tar && cd ../ - elif [[ ${model_name} =~ "ch_ppocr_mobile_v2.0" ]]; then + elif [[ ${model_name} =~ "ch_ppocr_mobile_v2_0" ]]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_det_infer.tar && tar xf ch_ppocr_mobile_v2.0_rec_infer.tar && cd ../ - elif [[ ${model_name} =~ "ch_ppocr_server_v2.0" ]]; then + elif [[ ${model_name} =~ "ch_ppocr_server_v2_0" ]]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_server_v2.0_det_infer.tar && tar xf ch_ppocr_server_v2.0_rec_infer.tar && cd ../ @@ -650,11 +664,11 @@ if [ ${MODE} = "paddle2onnx_infer" ];then ${python_name} -m pip install paddle2onnx ${python_name} -m pip install onnxruntime # wget model - if [[ ${model_name} =~ "ch_ppocr_mobile_v2.0" ]]; then + if [[ ${model_name} =~ "ch_ppocr_mobile_v2_0" ]]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_det_infer.tar && tar xf ch_ppocr_mobile_v2.0_rec_infer.tar && cd ../ - elif [[ ${model_name} =~ "ch_ppocr_server_v2.0" ]]; then + elif [[ ${model_name} =~ "ch_ppocr_server_v2_0" ]]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_server_v2.0_det_infer.tar && tar xf ch_ppocr_server_v2.0_rec_infer.tar && cd ../ From 5afce66fc2a05ca51d729e12c5ab661aefb44af9 Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Tue, 2 Aug 2022 19:43:36 +0800 Subject: [PATCH 03/25] rename 2.0 to 2_0 --- test_tipc/configs/ch_PP-OCRv3_rec/train_infer_python.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/test_tipc/configs/ch_PP-OCRv3_rec/train_infer_python.txt b/test_tipc/configs/ch_PP-OCRv3_rec/train_infer_python.txt index 84cceb6682..59fc1bd416 100644 --- a/test_tipc/configs/ch_PP-OCRv3_rec/train_infer_python.txt +++ b/test_tipc/configs/ch_PP-OCRv3_rec/train_infer_python.txt @@ -56,4 +56,5 @@ batch_size:64 fp_items:fp32|fp16 epoch:1 --profiler_options:batch_range=[10,20];state=GPU;tracer_option=Default;profile_path=model.profile +flags:FLAGS_eager_delete_tensor_gb=0.0;FLAGS_fraction_of_gpu_memory_to_use=0.98;FLAGS_conv_workspace_size_limit=4096 From 28daed6f3cd87dbeb41b471b0410aababc9d3fe5 Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Tue, 2 Aug 2022 19:54:28 +0800 Subject: [PATCH 04/25] fix bug --- .../model_linux_gpu_normal_normal_infer_python_mac_cpu.txt | 2 +- ...del_linux_gpu_normal_normal_infer_python_windows_gpu_cpu.txt | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_mac_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_mac_cpu.txt index 2d8a680b72..ea74b7fe76 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_mac_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_mac_cpu.txt @@ -1,5 +1,5 @@ ===========================kl_quant_params=========================== -infer_model:./inference/ch_ppocr_mobile_v2_0_det_infer/ +infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ infer_export:deploy/slim/quantization/quant_kl.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o infer_quant:True inference:tools/infer/predict_det.py diff --git a/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_windows_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_windows_gpu_cpu.txt index 2d8a680b72..ea74b7fe76 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_windows_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_det_KL/model_linux_gpu_normal_normal_infer_python_windows_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================kl_quant_params=========================== -infer_model:./inference/ch_ppocr_mobile_v2_0_det_infer/ +infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ infer_export:deploy/slim/quantization/quant_kl.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o infer_quant:True inference:tools/infer/predict_det.py From e7b160113d9c9337c608cee8cf44289f3bc3db5d Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Tue, 2 Aug 2022 20:02:25 +0800 Subject: [PATCH 05/25] fix fp16 bug --- tools/train.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tools/train.py b/tools/train.py index 309d4bb9e6..dc8cae8a63 100755 --- a/tools/train.py +++ b/tools/train.py @@ -119,9 +119,6 @@ def main(config, device, logger, vdl_writer): config['Loss']['ignore_index'] = char_num - 1 model = build_model(config['Architecture']) - if config['Global']['distributed']: - model = paddle.DataParallel(model) - model = apply_to_static(model, config, logger) # build loss @@ -157,10 +154,13 @@ def main(config, device, logger, vdl_writer): scaler = paddle.amp.GradScaler( init_loss_scaling=scale_loss, use_dynamic_loss_scaling=use_dynamic_loss_scaling) - model, optimizer = paddle.amp.decorate(models=model, optimizers=optimizer, level='O2', master_weight=True) + model, optimizer = paddle.amp.decorate( + models=model, optimizers=optimizer, level='O2', master_weight=True) else: scaler = None + if config['Global']['distributed']: + model = paddle.DataParallel(model) # start train program.train(config, train_dataloader, valid_dataloader, device, model, loss_class, optimizer, lr_scheduler, post_process_class, From 0cd06d9cf95c365adf7fb30d9e79d39572a9398a Mon Sep 17 00:00:00 2001 From: MissPenguin Date: Wed, 3 Aug 2022 10:35:21 +0800 Subject: [PATCH 06/25] Update detection_en.md --- doc/doc_en/detection_en.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/doc_en/detection_en.md b/doc/doc_en/detection_en.md index f85bf585cb..c215e1a466 100644 --- a/doc/doc_en/detection_en.md +++ b/doc/doc_en/detection_en.md @@ -51,7 +51,7 @@ python3 tools/train.py -c configs/det/det_mv3_db.yml \ -o Global.pretrained_model=./pretrain_models/MobileNetV3_large_x0_5_pretrained ``` -In the above instruction, use `-c` to select the training to use the `configs/det/det_db_mv3.yml` configuration file. +In the above instruction, use `-c` to select the training to use the `configs/det/det_mv3_db.yml` configuration file. For a detailed explanation of the configuration file, please refer to [config](./config_en.md). You can also use `-o` to change the training parameters without modifying the yml file. For example, adjust the training learning rate to 0.0001 From 0bec6bf4545e0b27f6d80f963906d8f8b0fb25fc Mon Sep 17 00:00:00 2001 From: MissPenguin Date: Wed, 3 Aug 2022 10:35:47 +0800 Subject: [PATCH 07/25] Update detection.md --- doc/doc_ch/detection.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/doc_ch/detection.md b/doc/doc_ch/detection.md index 2cf0732219..eba5213501 100644 --- a/doc/doc_ch/detection.md +++ b/doc/doc_ch/detection.md @@ -65,7 +65,7 @@ python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/ ``` -上述指令中，通过-c 选择训练使用configs/det/det_db_mv3.yml配置文件。 +上述指令中，通过-c 选择训练使用configs/det/det_mv3_db.yml配置文件。 有关配置文件的详细解释，请参考[链接](./config.md)。 您也可以通过-o参数在不需要修改yml文件的情况下，改变训练的参数，比如，调整训练的学习率为0.0001 From 56ada55d0848032952e47bd8e4ef34afccc2500b Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Wed, 3 Aug 2022 11:53:01 +0800 Subject: [PATCH 08/25] add layoutxlm --- .../train_infer_python.txt | 4 +- .../configs/layoutxlm/train_infer_python.txt | 59 +++++++++++++++++++ test_tipc/docs/benchmark_train.md | 1 + test_tipc/prepare.sh | 20 +++++++ 4 files changed, 82 insertions(+), 2 deletions(-) create mode 100644 test_tipc/configs/layoutxlm/train_infer_python.txt diff --git a/test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/train_infer_python.txt b/test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/train_infer_python.txt index cf176b3cff..cd00c7493a 100644 --- a/test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/train_infer_python.txt +++ b/test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/train_infer_python.txt @@ -27,7 +27,7 @@ null:null ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/det_r50_vd_sast_totaltext_v2.0/det_r50_vd_sast_totaltext.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/det_r50_vd_sast_totaltext.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null inference_dir:null train_model:./inference/det_r50_vd_sast_totaltext_v2.0_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/det_r50_vd_sast_totaltext_v2.0/det_r50_vd_sast_totaltext.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/det_r50_vd_sast_totaltext.yml -o infer_quant:False inference:tools/infer/predict_det.py --use_gpu:True|False diff --git a/test_tipc/configs/layoutxlm/train_infer_python.txt b/test_tipc/configs/layoutxlm/train_infer_python.txt new file mode 100644 index 0000000000..32b623a8dc --- /dev/null +++ b/test_tipc/configs/layoutxlm/train_infer_python.txt @@ -0,0 +1,59 @@ +===========================train_params=========================== +model_name:layoutxlm +python:python3.7 +gpu_list:0|0,1 +Global.use_gpu:True|True +Global.auto_cast:fp32 +Global.epoch_num:lite_train_lite_infer=1|whole_train_whole_infer=17 +Global.save_model_dir:./output/ +Train.loader.batch_size_per_card:lite_train_lite_infer=8|whole_train_whole_infer=8 +Architecture.Backbone.checkpoints:null +train_model_name:latest +train_infer_img_dir:ppstructure/docs/vqa/input/zh_val_42.jpg +null:null +## +trainer:norm_train +norm_train:tools/train.py -c configs/vqa/ser/layoutxlm_xfund_zh.yml -o Global.print_batch_step=1 Global.eval_batch_step=[1000,1000] Train.loader.shuffle=false +pact_train:null +fpgm_train:null +distill_train:null +null:null +null:null +## +===========================eval_params=========================== +eval:null +null:null +## +===========================infer_params=========================== +Global.save_inference_dir:./output/ +Architecture.Backbone.checkpoints: +norm_export:tools/export_model.py -c configs/vqa/ser/layoutxlm_xfund_zh.yml -o +quant_export: +fpgm_export: +distill_export:null +export1:null +export2:null +## +infer_model:null +infer_export:null +infer_quant:False +inference:ppstructure/vqa/predict_vqa_token_ser.py --vqa_algorithm=LayoutXLM --ser_dict_path=train_data/XFUND/class_list_xfun.txt --output=output +--use_gpu:True|False +--enable_mkldnn:False +--cpu_threads:6 +--rec_batch_num:1 +--use_tensorrt:False +--precision:fp32 +--ser_model_dir: +--image_dir:./ppstructure/docs/vqa/input/zh_val_42.jpg +null:null +--benchmark:False +null:null +===========================infer_benchmark_params========================== +random_infer_input:[{float32,[3,224,224]}] +===========================train_benchmark_params========================== +batch_size:4 +fp_items:fp32|fp16 +epoch:3 +--profiler_options:batch_range=[10,20];state=GPU;tracer_option=Default;profile_path=model.profile +flags:FLAGS_eager_delete_tensor_gb=0.0;FLAGS_fraction_of_gpu_memory_to_use=0.98 diff --git a/test_tipc/docs/benchmark_train.md b/test_tipc/docs/benchmark_train.md index 0196edc247..9076ccc40c 100644 --- a/test_tipc/docs/benchmark_train.md +++ b/test_tipc/docs/benchmark_train.md @@ -69,6 +69,7 @@ train_log/ | det_r50_vd_east_v2.0 |[config](../configs/det_r50_vd_east_v2.0/train_infer_python.txt) | 42.485 | 42.624 / 42.663 / 42.561 |0.00239083 | 67.61 |67.825/ 68.299/ 68.51| 0.00999854 | 10,000| 2,000| | det_r50_vd_pse_v2.0 |[config](../configs/det_r50_vd_pse_v2.0/train_infer_python.txt) | 16.455 | 16.517 / 16.555 / 16.353 |0.012201752 | 27.02 |27.288 / 27.152 / 27.408| 0.009340339 | 10,000| 2,000| | rec_mv3_none_bilstm_ctc_v2.0 |[config](../configs/rec_mv3_none_bilstm_ctc_v2.0/train_infer_python.txt) | 2288.358 | 2291.906 / 2293.725 / 2290.05 |0.001602197 | 2336.17 |2327.042 / 2328.093 / 2344.915| 0.007622025 | 600,000| 160,000| +| layoutxlm |[config](../configs/layoutxlm/train_infer_python.txt) | 18.001 | 18.114 / 18.107 / 18.307 |0.010924783 | 21.982 | 21.507 / 21.116 / 21.406| 0.018180127 | 1490 | 1490| | PP-Structure-table |[config](../configs/en_table_structure/train_infer_python.txt) | 14.151 | 14.077 / 14.23 / 14.25 |0.012140351 | 16.285 | 16.595 / 16.878 / 16.531 | 0.020559308 | 20,000| 5,000| | det_r50_dcn_fce_ctw_v2.0 |[config](../configs/det_r50_dcn_fce_ctw_v2.0/train_infer_python.txt) | 14.057 | 14.029 / 14.02 / 14.014 |0.001069214 | 18.298 |18.411 / 18.376 / 18.331| 0.004345228 | 10,000| 2,000| | ch_PP-OCRv3_det |[config](../configs/ch_PP-OCRv3_det/train_infer_python.txt) | 8.622 | 8.431 / 8.423 / 8.479|0.006604552 | 14.203 |14.346 14.468 14.23| 0.016450097 | 10,000| 2,000| diff --git a/test_tipc/prepare.sh b/test_tipc/prepare.sh index 88154d6187..35b9c73a06 100644 --- a/test_tipc/prepare.sh +++ b/test_tipc/prepare.sh @@ -106,6 +106,19 @@ if [ ${MODE} = "benchmark_train" ];then ln -s ./icdar2015_benckmark ./icdar2015 cd ../ fi + if [ ${model_name} == "layoutxlm" ]; then + pip install -r ppstructure/vqa/requirements.txt + pip install paddlenlp>=2.3.5 + wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/ppstructure/dataset/XFUND.tar --no-check-certificate + cd ./train_data/ && tar xf XFUND.tar + # expand gt.txt 10 times + cd XFUND/zh_train + for i in `seq 10`;do cp train.json dup$i.txt;done + cat dup* > train.json && rm -rf dup* + cd ../../ + + cd ../ + fi fi if [ ${MODE} = "lite_train_lite_infer" ];then @@ -207,6 +220,13 @@ if [ ${MODE} = "lite_train_lite_infer" ];then wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/rec_r32_gaspin_bilstm_att_train.tar --no-check-certificate cd ./pretrain_models/ && tar xf rec_r32_gaspin_bilstm_att_train.tar && cd ../ fi + if [ ${model_name} == "layoutxlm" ]; then + pip install -r ppstructure/vqa/requirements.txt + pip install paddlenlp>=2.3.5 + wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/ppstructure/dataset/XFUND.tar --no-check-certificate + cd ./train_data/ && tar xf XFUND.tar + cd ../ + fi elif [ ${MODE} = "whole_train_whole_infer" ];then wget -nc -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/MobileNetV3_large_x0_5_pretrained.pdparams --no-check-certificate From f6f7e3a3e55efa6ee8a7bd82bc59d7d2a9adb197 Mon Sep 17 00:00:00 2001 From: andyjpaddle Date: Wed, 3 Aug 2022 06:00:54 +0000 Subject: [PATCH 09/25] add handwrite case --- applications/手写文字识别.md | 249 +++++++++++++++++++++++++++++++++++ 1 file changed, 249 insertions(+) create mode 100644 applications/手写文字识别.md diff --git a/applications/手写文字识别.md b/applications/手写文字识别.md new file mode 100644 index 0000000000..1e78b142d8 --- /dev/null +++ b/applications/手写文字识别.md @@ -0,0 +1,249 @@ +# 基于PP-OCRv3的手写文字识别 + +- [1. 项目背景及意义](#1-项目背景及意义) +- [2. 项目内容](#2-项目内容) +- [3. PP-OCRv3识别算法介绍](#3-PP-OCRv3识别算法介绍) +- [4. 安装环境](#4-安装环境) +- [5. 数据准备](#5-数据准备) +- [6. 模型训练](#6-模型训练) + - [6.1 下载预训练模型](#61-下载预训练模型) + - [6.2 修改配置文件](#62-修改配置文件) + - [6.3 开始训练](#63-开始训练) +- [7. 模型评估](#7-模型评估) +- [8. 模型导出推理](#8-模型导出推理) + - [8.1 模型导出](#81-模型导出) + - [8.2 模型推理](#82-模型推理) + + +## 1. 项目背景及意义 +目前光学字符识别(OCR)技术在我们的生活当中被广泛使用，但是大多数模型在通用场景下的准确性还有待提高。针对于此我们借助飞桨提供的PaddleOCR套件较容易的实现了在垂类场景下的应用。手写体在日常生活中较为常见，然而手写体的识别却存在着很大的挑战，因为每个人的手写字体风格不一样，这对于视觉模型来说还是相当有挑战的。因此训练一个手写体识别模型具有很好的现实意义。 + +## 2. 项目内容 +本项目基于PaddleOCR套件，以PP-OCRv3识别模型为基础，针对手写文字识别场景进行优化。 + +Aistudio项目链接：[OCR手写文字识别](https://aistudio.baidu.com/aistudio/projectdetail/4330587) + +## 3. PP-OCRv3识别算法介绍 +PP-OCRv3的识别模块是基于文本识别算法[SVTR](https://arxiv.org/abs/2205.00159)优化。SVTR不再采用RNN结构，通过引入Transformers结构更加有效地挖掘文本行图像的上下文信息，从而提升文本识别能力。如下图所示，PP-OCRv3采用了6个优化策略。 + +![v3_rec](https://ai-studio-static-online.cdn.bcebos.com/d4f5344b5b854d50be738671598a89a45689c6704c4d481fb904dd7cf72f2a1a) + +优化策略汇总如下： + +* SVTR_LCNet：轻量级文本识别网络 +* GTC：Attention指导CTC训练策略 +* TextConAug：挖掘文字上下文信息的数据增广策略 +* TextRotNet：自监督的预训练模型 +* UDML：联合互学习策略 +* UIM：无标注数据挖掘方案 + +详细优化策略描述请参考[PP-OCRv3优化策略](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.5/doc/doc_ch/PP-OCRv3_introduction.md#3-%E8%AF%86%E5%88%AB%E4%BC%98%E5%8C%96) + +## 4. 安装环境 + + +```python +# 首先git官方的PaddleOCR项目，安装需要的依赖 +git clone https://github.com/PaddlePaddle/PaddleOCR.git +cd PaddleOCR +pip install -r requirements.txt +``` + +## 5. 数据准备 +本项目使用公开的手写文本识别数据集，包含Chinese OCR, 中科院自动化研究所-手写中文数据集[CASIA-HWDB2.x](http://www.nlpr.ia.ac.cn/databases/handwriting/Download.html)，以及由中科院手写数据和网上开源数据合并组合的[数据集](https://aistudio.baidu.com/aistudio/datasetdetail/102884/0)等，该项目已经挂载处理好的数据集，可直接下载使用进行训练。 + + +```python +下载并解压数据 +tar -xf hw_data.tar +``` + +## 6. 模型训练 +### 6.1 下载预训练模型 +首先需要下载我们需要的PP-OCRv3识别预训练模型，更多选择请自行选择其他的[文字识别模型](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.5/doc/doc_ch/models_list.md#2-%E6%96%87%E6%9C%AC%E8%AF%86%E5%88%AB%E6%A8%A1%E5%9E%8B) + + +```python +# 使用该指令下载需要的预训练模型 +wget -P ./pretrained_models/ https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_train.tar +# 解压预训练模型文件 +tar -xf ./pretrained_models/ch_PP-OCRv3_rec_train.tar -C pretrained_models +``` + +### 6.2 修改配置文件 +我们使用`configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml`，主要修改训练轮数和学习率参相关参数，设置预训练模型路径，设置数据集路径。 另外，batch_size可根据自己机器显存大小进行调整。 具体修改如下几个地方： + +``` + epoch_num: 100 # 训练epoch数 + save_model_dir: ./output/ch_PP-OCR_v3_rec + save_epoch_step: 10 + eval_batch_step: [0, 100] # 评估间隔，每隔100step评估一次 + pretrained_model: ./pretrained_models/ch_PP-OCRv3_rec_train/best_accuracy # 预训练模型路径 + + + lr: + name: Cosine # 修改学习率衰减策略为Cosine + learning_rate: 0.0001 # 修改fine-tune的学习率 + warmup_epoch: 2 # 修改warmup轮数 + +Train: + dataset: + name: SimpleDataSet + data_dir: ./train_data # 训练集图片路径 + ext_op_transform_idx: 1 + label_file_list: + - ./train_data/chineseocr-data/rec_hand_line_all_label_train.txt # 训练集标签 + - ./train_data/handwrite/HWDB2.0Train_label.txt + - ./train_data/handwrite/HWDB2.1Train_label.txt + - ./train_data/handwrite/HWDB2.2Train_label.txt + - ./train_data/handwrite/hwdb_ic13/handwriting_hwdb_train_labels.txt + - ./train_data/handwrite/HW_Chinese/train_hw.txt + ratio_list: + - 0.1 + - 1.0 + - 1.0 + - 1.0 + - 0.02 + - 1.0 + loader: + shuffle: true + batch_size_per_card: 64 + drop_last: true + num_workers: 4 +Eval: + dataset: + name: SimpleDataSet + data_dir: ./train_data # 测试集图片路径 + label_file_list: + - ./train_data/chineseocr-data/rec_hand_line_all_label_val.txt # 测试集标签 + - ./train_data/handwrite/HWDB2.0Test_label.txt + - ./train_data/handwrite/HWDB2.1Test_label.txt + - ./train_data/handwrite/HWDB2.2Test_label.txt + - ./train_data/handwrite/hwdb_ic13/handwriting_hwdb_val_labels.txt + - ./train_data/handwrite/HW_Chinese/test_hw.txt + loader: + shuffle: false + drop_last: false + batch_size_per_card: 64 + num_workers: 4 +``` +由于数据集大多是长文本，因此需要**注释**掉下面的数据增广策略，以便训练出更好的模型。 +``` +- RecConAug: + prob: 0.5 + ext_data_num: 2 + image_shape: [48, 320, 3] +``` + + +### 6.3 开始训练 +我们使用上面修改好的配置文件`configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml`，预训练模型，数据集路径，学习率，训练轮数等都已经设置完毕后，可以使用下面命令开始训练。 + + +```python +# 开始训练识别模型 +python tools/train.py -c configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml + +``` + +## 7. 模型评估 +在训练之前，我们可以直接使用下面命令来评估预训练模型的效果: + + + +```python +# 评估预训练模型 +python tools/eval.py -c configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml -o Global.pretrained_model="./pretrained_models/ch_PP-OCRv3_rec_train/best_accuracy" +``` +``` +[2022/07/14 10:46:22] ppocr INFO: load pretrain successful from ./pretrained_models/ch_PP-OCRv3_rec_train/best_accuracy +eval model:: 100%|████████████████████████████| 687/687 [03:29<00:00, 3.27it/s] +[2022/07/14 10:49:52] ppocr INFO: metric eval *************** +[2022/07/14 10:49:52] ppocr INFO: acc:0.03724954461811258 +[2022/07/14 10:49:52] ppocr INFO: norm_edit_dis:0.4859541065843199 +[2022/07/14 10:49:52] ppocr INFO: Teacher_acc:0.0371584699368947 +[2022/07/14 10:49:52] ppocr INFO: Teacher_norm_edit_dis:0.48718814890536477 +[2022/07/14 10:49:52] ppocr INFO: fps:947.8562684823883 +``` + +可以看出，直接加载预训练模型进行评估，效果较差，因为预训练模型并不是基于手写文字进行单独训练的，所以我们需要基于预训练模型进行finetune。 +训练完成后，可以进行测试评估，评估命令如下： + + + +```python +# 评估finetune效果 +python tools/eval.py -c configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml -o Global.pretrained_model="./output/ch_PP-OCR_v3_rec/best_accuracy" + +``` + +评估结果如下，可以看出识别准确率为54.3%。 +``` +[2022/07/14 10:54:06] ppocr INFO: metric eval *************** +[2022/07/14 10:54:06] ppocr INFO: acc:0.5430100180913 +[2022/07/14 10:54:06] ppocr INFO: norm_edit_dis:0.9203322593158589 +[2022/07/14 10:54:06] ppocr INFO: Teacher_acc:0.5401183969626324 +[2022/07/14 10:54:06] ppocr INFO: Teacher_norm_edit_dis:0.919827504507755 +[2022/07/14 10:54:06] ppocr INFO: fps:928.948733797251 +``` + +如需获取已训练模型，请扫码填写问卷，加入PaddleOCR官方交流群获取全部OCR垂类模型下载链接、《动手学OCR》电子书等全套OCR学习资料🎁 +
+ +
+将下载或训练完成的模型放置在对应目录下即可完成模型推理。 + +## 8. 模型导出推理 +训练完成后，可以将训练模型转换成inference模型。inference 模型会额外保存模型的结构信息，在预测部署、加速推理上性能优越，灵活方便，适合于实际系统集成。 + + +### 8.1 模型导出 +导出命令如下： + + + +```python +# 转化为推理模型 +python tools/export_model.py -c configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml -o Global.pretrained_model="./output/ch_PP-OCR_v3_rec/best_accuracy" Global.save_inference_dir="./inference/rec_ppocrv3/" + +``` + +### 8.2 模型推理 +导出模型后，可以使用如下命令进行推理预测: + + + +```python +# 推理预测 +python tools/infer/predict_rec.py --image_dir="train_data/handwrite/HWDB2.0Test_images/104-P16_4.jpg" --rec_model_dir="./inference/rec_ppocrv3/Student" +``` + +``` +[2022/07/14 10:55:56] ppocr INFO: In PP-OCRv3, rec_image_shape parameter defaults to '3, 48, 320', if you are using recognition model with PP-OCRv2 or an older version, please set --rec_image_shape='3,32,320 +[2022/07/14 10:55:58] ppocr INFO: Predicts of train_data/handwrite/HWDB2.0Test_images/104-P16_4.jpg:('品结构,差异化的多品牌渗透使欧莱雅确立了其在中国化妆', 0.9904912114143372) +``` + + +```python +# 可视化文字识别图片 +from PIL import Image +import matplotlib.pyplot as plt +import numpy as np +import os + + +img_path = 'train_data/handwrite/HWDB2.0Test_images/104-P16_4.jpg' + +def vis(img_path): + plt.figure() + image = Image.open(img_path) + plt.imshow(image) + plt.show() + # image = image.resize([208, 208]) + + +vis(img_path) +``` + + +![res](https://ai-studio-static-online.cdn.bcebos.com/ad7c02745491498d82e0ce95f4a274f9b3920b2f467646858709359b7af9d869) From d59c7f672bc7781a94d67fe60f252fca9ace945f Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Wed, 3 Aug 2022 14:25:06 +0800 Subject: [PATCH 10/25] rename 2.0 to 2_0 --- .../ch_ppocr_mobile_v2_0_rec_FPGM/train_infer_python.txt | 4 ++-- ...in_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt | 4 ++-- .../ch_ppocr_server_v2_0_det/train_infer_python.txt | 6 +++--- ..._linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt | 6 +++--- ...in_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt | 6 +++--- .../ch_ppocr_server_v2_0_rec/train_infer_python.txt | 8 ++++---- ..._linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt | 8 ++++---- ...in_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt | 8 ++++---- .../configs/det_mv3_east_v2_0/train_infer_python.txt | 6 +++--- test_tipc/configs/det_mv3_pse_v2_0/train_infer_python.txt | 6 +++--- .../det_r50_dcn_fce_ctw_v2_0/train_infer_python.txt | 6 +++--- .../det_r50_vd_sast_icdar15_v2_0/train_infer_python.txt | 6 +++--- .../det_r50_vd_sast_totaltext_v2_0/train_infer_python.txt | 2 +- .../rec_mv3_none_bilstm_ctc_v2_0/train_infer_python.txt | 8 ++++---- .../rec_mv3_none_none_ctc_v2_0/train_infer_python.txt | 8 ++++---- .../rec_mv3_tps_bilstm_att_v2_0/train_infer_python.txt | 8 ++++---- .../rec_mv3_tps_bilstm_ctc_v2_0/train_infer_python.txt | 8 ++++---- .../train_infer_python.txt | 8 ++++---- .../rec_r34_vd_none_none_ctc_v2_0/train_infer_python.txt | 8 ++++---- .../rec_r34_vd_tps_bilstm_att_v2_0/train_infer_python.txt | 8 ++++---- .../rec_r34_vd_tps_bilstm_ctc_v2_0/train_infer_python.txt | 8 ++++---- 21 files changed, 70 insertions(+), 70 deletions(-) diff --git a/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_infer_python.txt index b2fb028823..94c9503103 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_infer_python.txt @@ -15,7 +15,7 @@ null:null trainer:fpgm_train norm_train:null pact_train:null -fpgm_train:deploy/slim/prune/sensitivity_anal.py -c test_tipc/configs/ch_ppocr_mobile_v2.0_rec_FPGM/rec_chinese_lite_train_v2.0.yml -o Global.pretrained_model=./pretrain_models/ch_ppocr_mobile_v2.0_rec_train/best_accuracy +fpgm_train:deploy/slim/prune/sensitivity_anal.py -c test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/rec_chinese_lite_train_v2.0.yml -o Global.pretrained_model=./pretrain_models/ch_ppocr_mobile_v2.0_rec_train/best_accuracy distill_train:null null:null null:null @@ -29,7 +29,7 @@ Global.save_inference_dir:./output/ Global.checkpoints: norm_export:null quant_export:null -fpgm_export:deploy/slim/prune/export_prune_model.py -c test_tipc/configs/ch_ppocr_mobile_v2.0_rec_FPGM/rec_chinese_lite_train_v2.0.yml -o +fpgm_export:deploy/slim/prune/export_prune_model.py -c test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/rec_chinese_lite_train_v2.0.yml -o distill_export:null export1:null export2:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt index 509240f6c5..71555865ad 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt @@ -15,7 +15,7 @@ null:null trainer:fpgm_train norm_train:null pact_train:null -fpgm_train:deploy/slim/prune/sensitivity_anal.py -c test_tipc/configs/ch_ppocr_mobile_v2.0_rec_FPGM/rec_chinese_lite_train_v2.0.yml -o Global.pretrained_model=./pretrain_models/ch_ppocr_mobile_v2.0_rec_train/best_accuracy +fpgm_train:deploy/slim/prune/sensitivity_anal.py -c test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/rec_chinese_lite_train_v2.0.yml -o Global.pretrained_model=./pretrain_models/ch_ppocr_mobile_v2.0_rec_train/best_accuracy distill_train:null null:null null:null @@ -29,7 +29,7 @@ Global.save_inference_dir:./output/ Global.checkpoints: norm_export:null quant_export:null -fpgm_export:deploy/slim/prune/export_prune_model.py -c test_tipc/configs/ch_ppocr_mobile_v2.0_rec_FPGM/rec_chinese_lite_train_v2.0.yml -o +fpgm_export:deploy/slim/prune/export_prune_model.py -c test_tipc/configs/ch_ppocr_mobile_v2_0_rec_FPGM/rec_chinese_lite_train_v2.0.yml -o distill_export:null export1:null export2:null diff --git a/test_tipc/configs/ch_ppocr_server_v2_0_det/train_infer_python.txt b/test_tipc/configs/ch_ppocr_server_v2_0_det/train_infer_python.txt index 87e18ef7f3..90ed29f430 100644 --- a/test_tipc/configs/ch_ppocr_server_v2_0_det/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_det/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_server_v2.0_det/det_r50_vd_db.yml -o +norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_server_v2_0_det/det_r50_vd_db.yml -o quant_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/ch_ppocr_server_v2.0_det/det_r50_vd_db.yml -o +eval:tools/eval.py -c test_tipc/configs/ch_ppocr_server_v2_0_det/det_r50_vd_db.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2.0_det/det_r50_vd_db.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2_0_det/det_r50_vd_db.yml -o quant_export:null fpgm_export:null distill_export:null diff --git a/test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt index d1c7cf1d9e..f398078fc5 100644 --- a/test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_server_v2.0_det/det_r50_vd_db.yml -o +norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_server_v2_0_det/det_r50_vd_db.yml -o quant_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/ch_ppocr_server_v2.0_det/det_r50_vd_db.yml -o +eval:tools/eval.py -c test_tipc/configs/ch_ppocr_server_v2_0_det/det_r50_vd_db.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2.0_det/det_r50_vd_db.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2_0_det/det_r50_vd_db.yml -o quant_export:null fpgm_export:null distill_export:null diff --git a/test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt index 2802406b87..7a2d0a53c1 100644 --- a/test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_det/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_server_v2.0_det/det_r50_vd_db.yml -o +norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_server_v2_0_det/det_r50_vd_db.yml -o quant_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/ch_ppocr_server_v2.0_det/det_r50_vd_db.yml -o +eval:tools/eval.py -c test_tipc/configs/ch_ppocr_server_v2_0_det/det_r50_vd_db.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2.0_det/det_r50_vd_db.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2_0_det/det_r50_vd_db.yml -o quant_export:null fpgm_export:null distill_export:null diff --git a/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_infer_python.txt b/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_infer_python.txt index e7cde1334a..b9a1ae4984 100644 --- a/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./inference/rec_inference null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml -o Global.print_batch_step=4 Train.loader.shuffle=false +norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml -o Global.print_batch_step=4 Train.loader.shuffle=false pact_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml -o +eval:tools/eval.py -c test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/ch_ppocr_server_v2.0_rec_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml -o infer_quant:False inference:tools/infer/predict_rec.py --use_gpu:True|False diff --git a/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt index 99b864d1b0..d5f57aef9f 100644 --- a/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./inference/rec_inference null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml -o +norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml -o pact_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml -o +eval:tools/eval.py -c test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/ch_ppocr_server_v2.0_rec_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml -o infer_quant:False inference:tools/infer/predict_rec.py --use_gpu:False diff --git a/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt index 63aa07dd18..20eb10b8e6 100644 --- a/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_server_v2_0_rec/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./inference/rec_inference null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml -o +norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml -o pact_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml -o +eval:tools/eval.py -c test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/ch_ppocr_server_v2.0_rec_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2.0_rec/rec_icdar15_train.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_server_v2_0_rec/rec_icdar15_train.yml -o infer_quant:False inference:tools/infer/predict_rec.py --use_gpu:True|False diff --git a/test_tipc/configs/det_mv3_east_v2_0/train_infer_python.txt b/test_tipc/configs/det_mv3_east_v2_0/train_infer_python.txt index 2818b5041c..9c6d9660d5 100644 --- a/test_tipc/configs/det_mv3_east_v2_0/train_infer_python.txt +++ b/test_tipc/configs/det_mv3_east_v2_0/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/det_mv3_east_v2.0/det_mv3_east.yml -o +norm_train:tools/train.py -c test_tipc/configs/det_mv3_east_v2_0/det_mv3_east.yml -o pact_train:null fpgm_train:null distill_train:null @@ -27,7 +27,7 @@ null:null ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/det_mv3_east_v2.0/det_mv3_east.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/det_mv3_east_v2_0/det_mv3_east.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/det_mv3_east_v2.0_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/det_mv3_east_v2.0/det_mv3_east.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/det_mv3_east_v2_0/det_mv3_east.yml -o infer_quant:False inference:tools/infer/predict_det.py --use_gpu:True|False diff --git a/test_tipc/configs/det_mv3_pse_v2_0/train_infer_python.txt b/test_tipc/configs/det_mv3_pse_v2_0/train_infer_python.txt index 1fb4724ea4..525fdc7d4b 100644 --- a/test_tipc/configs/det_mv3_pse_v2_0/train_infer_python.txt +++ b/test_tipc/configs/det_mv3_pse_v2_0/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/det_mv3_pse_v2.0/det_mv3_pse.yml -o +norm_train:tools/train.py -c test_tipc/configs/det_mv3_pse_v2_0/det_mv3_pse.yml -o pact_train:null fpgm_train:null distill_train:null @@ -27,7 +27,7 @@ null:null ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/det_mv3_pse_v2.0/det_mv3_pse.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/det_mv3_pse_v2_0/det_mv3_pse.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/det_mv3_pse_v2.0_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/det_mv3_pse_v2.0/det_mv3_pse.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/det_mv3_pse_v2_0/det_mv3_pse.yml -o infer_quant:False inference:tools/infer/predict_det.py --use_gpu:True|False diff --git a/test_tipc/configs/det_r50_dcn_fce_ctw_v2_0/train_infer_python.txt b/test_tipc/configs/det_r50_dcn_fce_ctw_v2_0/train_infer_python.txt index 4454d018ac..92ded19d67 100644 --- a/test_tipc/configs/det_r50_dcn_fce_ctw_v2_0/train_infer_python.txt +++ b/test_tipc/configs/det_r50_dcn_fce_ctw_v2_0/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/det_r50_dcn_fce_ctw_v2.0/det_r50_vd_dcn_fce_ctw.yml -o Global.print_batch_step=1 Train.loader.shuffle=false +norm_train:tools/train.py -c test_tipc/configs/det_r50_dcn_fce_ctw_v2_0/det_r50_vd_dcn_fce_ctw.yml -o Global.print_batch_step=1 Train.loader.shuffle=false pact_train:null fpgm_train:null distill_train:null @@ -27,7 +27,7 @@ null:null ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/det_r50_dcn_fce_ctw_v2.0/det_r50_vd_dcn_fce_ctw.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/det_r50_dcn_fce_ctw_v2_0/det_r50_vd_dcn_fce_ctw.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/det_r50_dcn_fce_ctw_v2.0_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/det_r50_dcn_fce_ctw_v2.0/det_r50_vd_dcn_fce_ctw.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/det_r50_dcn_fce_ctw_v2_0/det_r50_vd_dcn_fce_ctw.yml -o infer_quant:False inference:tools/infer/predict_det.py --use_gpu:True|False diff --git a/test_tipc/configs/det_r50_vd_sast_icdar15_v2_0/train_infer_python.txt b/test_tipc/configs/det_r50_vd_sast_icdar15_v2_0/train_infer_python.txt index 718aff271d..b01f1925b4 100644 --- a/test_tipc/configs/det_r50_vd_sast_icdar15_v2_0/train_infer_python.txt +++ b/test_tipc/configs/det_r50_vd_sast_icdar15_v2_0/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/det_r50_vd_sast_icdar15_v2.0/det_r50_vd_sast_icdar2015.yml -o +norm_train:tools/train.py -c test_tipc/configs/det_r50_vd_sast_icdar15_v2_0/det_r50_vd_sast_icdar2015.yml -o pact_train:null fpgm_train:null distill_train:null @@ -27,7 +27,7 @@ null:null ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/det_r50_vd_sast_icdar15_v2.0/det_r50_vd_sast_icdar2015.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/det_r50_vd_sast_icdar15_v2_0/det_r50_vd_sast_icdar2015.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null inference_dir:null train_model:./inference/det_r50_vd_sast_icdar15_v2.0_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/det_r50_vd_sast_icdar15_v2.0/det_r50_vd_sast_icdar2015.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/det_r50_vd_sast_icdar15_v2_0/det_r50_vd_sast_icdar2015.yml -o infer_quant:False inference:tools/infer/predict_det.py --use_gpu:True|False diff --git a/test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/train_infer_python.txt b/test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/train_infer_python.txt index cd00c7493a..a47ad68030 100644 --- a/test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/train_infer_python.txt +++ b/test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/det_r50_vd_sast_totaltext_v2.0/det_r50_vd_sast_totaltext.yml -o Global.pretrained_model=./pretrain_models/ResNet50_vd_ssld_pretrained +norm_train:tools/train.py -c test_tipc/configs/det_r50_vd_sast_totaltext_v2_0/det_r50_vd_sast_totaltext.yml -o Global.pretrained_model=./pretrain_models/ResNet50_vd_ssld_pretrained pact_train:null fpgm_train:null distill_train:null diff --git a/test_tipc/configs/rec_mv3_none_bilstm_ctc_v2_0/train_infer_python.txt b/test_tipc/configs/rec_mv3_none_bilstm_ctc_v2_0/train_infer_python.txt index 9e1fb14a0a..db89b4c78d 100644 --- a/test_tipc/configs/rec_mv3_none_bilstm_ctc_v2_0/train_infer_python.txt +++ b/test_tipc/configs/rec_mv3_none_bilstm_ctc_v2_0/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./inference/rec_inference null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/rec_mv3_none_bilstm_ctc_v2.0/rec_icdar15_train.yml -o Global.print_batch_step=4 Train.loader.shuffle=false +norm_train:tools/train.py -c test_tipc/configs/rec_mv3_none_bilstm_ctc_v2_0/rec_icdar15_train.yml -o Global.print_batch_step=4 Train.loader.shuffle=false pact_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/rec_mv3_none_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +eval:tools/eval.py -c test_tipc/configs/rec_mv3_none_bilstm_ctc_v2_0/rec_icdar15_train.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/rec_mv3_none_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/rec_mv3_none_bilstm_ctc_v2_0/rec_icdar15_train.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/rec_mv3_none_bilstm_ctc_v2.0_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/rec_mv3_none_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/rec_mv3_none_bilstm_ctc_v2_0/rec_icdar15_train.yml -o infer_quant:False inference:tools/infer/predict_rec.py --rec_char_dict_path=./ppocr/utils/ic15_dict.txt --rec_image_shape="3,32,100" --use_gpu:True|False diff --git a/test_tipc/configs/rec_mv3_none_none_ctc_v2_0/train_infer_python.txt b/test_tipc/configs/rec_mv3_none_none_ctc_v2_0/train_infer_python.txt index 09fa408cab..003e91ff3d 100644 --- a/test_tipc/configs/rec_mv3_none_none_ctc_v2_0/train_infer_python.txt +++ b/test_tipc/configs/rec_mv3_none_none_ctc_v2_0/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./inference/rec_inference null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/rec_mv3_none_none_ctc_v2.0/rec_icdar15_train.yml -o +norm_train:tools/train.py -c test_tipc/configs/rec_mv3_none_none_ctc_v2_0/rec_icdar15_train.yml -o pact_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/rec_mv3_none_none_ctc_v2.0/rec_icdar15_train.yml -o +eval:tools/eval.py -c test_tipc/configs/rec_mv3_none_none_ctc_v2_0/rec_icdar15_train.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/rec_mv3_none_none_ctc_v2.0/rec_icdar15_train.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/rec_mv3_none_none_ctc_v2_0/rec_icdar15_train.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/rec_mv3_none_none_ctc_v2.0_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/rec_mv3_none_none_ctc_v2.0/rec_icdar15_train.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/rec_mv3_none_none_ctc_v2_0/rec_icdar15_train.yml -o infer_quant:False inference:tools/infer/predict_rec.py --rec_char_dict_path=./ppocr/utils/ic15_dict.txt --rec_image_shape="3,32,100" --use_gpu:True|False diff --git a/test_tipc/configs/rec_mv3_tps_bilstm_att_v2_0/train_infer_python.txt b/test_tipc/configs/rec_mv3_tps_bilstm_att_v2_0/train_infer_python.txt index 6a1d76810c..c7b416c833 100644 --- a/test_tipc/configs/rec_mv3_tps_bilstm_att_v2_0/train_infer_python.txt +++ b/test_tipc/configs/rec_mv3_tps_bilstm_att_v2_0/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./inference/rec_inference null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/rec_mv3_tps_bilstm_att_v2.0/rec_mv3_tps_bilstm_att.yml -o +norm_train:tools/train.py -c test_tipc/configs/rec_mv3_tps_bilstm_att_v2_0/rec_mv3_tps_bilstm_att.yml -o pact_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/rec_mv3_tps_bilstm_att_v2.0/rec_mv3_tps_bilstm_att.yml -o +eval:tools/eval.py -c test_tipc/configs/rec_mv3_tps_bilstm_att_v2_0/rec_mv3_tps_bilstm_att.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/rec_mv3_tps_bilstm_att_v2.0/rec_mv3_tps_bilstm_att.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/rec_mv3_tps_bilstm_att_v2_0/rec_mv3_tps_bilstm_att.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/rec_mv3_tps_bilstm_att_v2.0_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/rec_mv3_tps_bilstm_att_v2.0/rec_mv3_tps_bilstm_att.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/rec_mv3_tps_bilstm_att_v2_0/rec_mv3_tps_bilstm_att.yml -o infer_quant:False inference:tools/infer/predict_rec.py --rec_char_dict_path=./ppocr/utils/ic15_dict.txt --rec_image_shape="3,32,100" --rec_algorithm="RARE" --min_subgraph_size=5 --use_gpu:True|False diff --git a/test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2_0/train_infer_python.txt b/test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2_0/train_infer_python.txt index 1b6be1b04b..0c6e2d1da7 100644 --- a/test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2_0/train_infer_python.txt +++ b/test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2_0/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./inference/rec_inference null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +norm_train:tools/train.py -c test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2_0/rec_icdar15_train.yml -o pact_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +eval:tools/eval.py -c test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2_0/rec_icdar15_train.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2_0/rec_icdar15_train.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/rec_mv3_tps_bilstm_ctc_v2.0_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/rec_mv3_tps_bilstm_ctc_v2_0/rec_icdar15_train.yml -o infer_quant:False inference:tools/infer/predict_rec.py --rec_char_dict_path=./ppocr/utils/ic15_dict.txt --rec_image_shape="3,32,100" --rec_algorithm="StarNet" --use_gpu:True|False diff --git a/test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2_0/train_infer_python.txt b/test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2_0/train_infer_python.txt index 98953b31fa..07a6190b0e 100644 --- a/test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2_0/train_infer_python.txt +++ b/test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2_0/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./inference/rec_inference null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +norm_train:tools/train.py -c test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2_0/rec_icdar15_train.yml -o pact_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +eval:tools/eval.py -c test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2_0/rec_icdar15_train.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2_0/rec_icdar15_train.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/rec_r34_vd_none_bilstm_ctc_v2.0_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_none_bilstm_ctc_v2_0/rec_icdar15_train.yml -o infer_quant:False inference:tools/infer/predict_rec.py --rec_char_dict_path=./ppocr/utils/ic15_dict.txt --rec_image_shape="3,32,100" --use_gpu:True|False diff --git a/test_tipc/configs/rec_r34_vd_none_none_ctc_v2_0/train_infer_python.txt b/test_tipc/configs/rec_r34_vd_none_none_ctc_v2_0/train_infer_python.txt index 37e66e8266..145793aa47 100644 --- a/test_tipc/configs/rec_r34_vd_none_none_ctc_v2_0/train_infer_python.txt +++ b/test_tipc/configs/rec_r34_vd_none_none_ctc_v2_0/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./inference/rec_inference null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/rec_r34_vd_none_none_ctc_v2.0/rec_icdar15_train.yml -o +norm_train:tools/train.py -c test_tipc/configs/rec_r34_vd_none_none_ctc_v2_0/rec_icdar15_train.yml -o pact_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/rec_r34_vd_none_none_ctc_v2.0/rec_icdar15_train.yml -o +eval:tools/eval.py -c test_tipc/configs/rec_r34_vd_none_none_ctc_v2_0/rec_icdar15_train.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_none_none_ctc_v2.0/rec_icdar15_train.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_none_none_ctc_v2_0/rec_icdar15_train.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/rec_r34_vd_none_none_ctc_v2.0_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_none_none_ctc_v2.0/rec_icdar15_train.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_none_none_ctc_v2_0/rec_icdar15_train.yml -o infer_quant:False inference:tools/infer/predict_rec.py --rec_char_dict_path=./ppocr/utils/ic15_dict.txt --rec_image_shape="3,32,100" --use_gpu:True|False diff --git a/test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2_0/train_infer_python.txt b/test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2_0/train_infer_python.txt index 772db83770..759518a4a1 100644 --- a/test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2_0/train_infer_python.txt +++ b/test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2_0/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./inference/rec_inference null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2.0/rec_r34_vd_tps_bilstm_att.yml -o +norm_train:tools/train.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2_0/rec_r34_vd_tps_bilstm_att.yml -o pact_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2.0/rec_r34_vd_tps_bilstm_att.yml -o +eval:tools/eval.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2_0/rec_r34_vd_tps_bilstm_att.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2.0/rec_r34_vd_tps_bilstm_att.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2_0/rec_r34_vd_tps_bilstm_att.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/rec_r34_vd_tps_bilstm_att_v2.0_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2.0/rec_r34_vd_tps_bilstm_att.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_att_v2_0/rec_r34_vd_tps_bilstm_att.yml -o infer_quant:False inference:tools/infer/predict_rec.py --rec_char_dict_path=./ppocr/utils/ic15_dict.txt --rec_image_shape="3,32,100" --rec_algorithm="RARE" --min_subgraph_size=5 --use_gpu:True|False diff --git a/test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2_0/train_infer_python.txt b/test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2_0/train_infer_python.txt index bc1ea7e851..ecc898341c 100644 --- a/test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2_0/train_infer_python.txt +++ b/test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2_0/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./inference/rec_inference null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +norm_train:tools/train.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2_0/rec_icdar15_train.yml -o pact_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval:tools/eval.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +eval:tools/eval.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2_0/rec_icdar15_train.yml -o null:null ## ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.checkpoints: -norm_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2_0/rec_icdar15_train.yml -o quant_export:null fpgm_export:null distill_export:null @@ -35,7 +35,7 @@ export1:null export2:null ## train_model:./inference/rec_r34_vd_tps_bilstm_ctc_v2.0_train/best_accuracy -infer_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2.0/rec_icdar15_train.yml -o +infer_export:tools/export_model.py -c test_tipc/configs/rec_r34_vd_tps_bilstm_ctc_v2_0/rec_icdar15_train.yml -o infer_quant:False inference:tools/infer/predict_rec.py --rec_char_dict_path=./ppocr/utils/ic15_dict.txt --rec_image_shape="3,32,100" --rec_algorithm="StarNet" --use_gpu:True|False From f16ff481786b317a5e4b3a962f8bc01285a23b13 Mon Sep 17 00:00:00 2001 From: andyjpaddle Date: Wed, 3 Aug 2022 06:41:33 +0000 Subject: [PATCH 11/25] add example img --- applications/手写文字识别.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/applications/手写文字识别.md b/applications/手写文字识别.md index 1e78b142d8..4796355bb7 100644 --- a/applications/手写文字识别.md +++ b/applications/手写文字识别.md @@ -16,7 +16,9 @@ ## 1. 项目背景及意义 -目前光学字符识别(OCR)技术在我们的生活当中被广泛使用，但是大多数模型在通用场景下的准确性还有待提高。针对于此我们借助飞桨提供的PaddleOCR套件较容易的实现了在垂类场景下的应用。手写体在日常生活中较为常见，然而手写体的识别却存在着很大的挑战，因为每个人的手写字体风格不一样，这对于视觉模型来说还是相当有挑战的。因此训练一个手写体识别模型具有很好的现实意义。 +目前光学字符识别(OCR)技术在我们的生活当中被广泛使用，但是大多数模型在通用场景下的准确性还有待提高。针对于此我们借助飞桨提供的PaddleOCR套件较容易的实现了在垂类场景下的应用。手写体在日常生活中较为常见，然而手写体的识别却存在着很大的挑战，因为每个人的手写字体风格不一样，这对于视觉模型来说还是相当有挑战的。因此训练一个手写体识别模型具有很好的现实意义。下面给出一些手写体的示例图： + +![example](https://ai-studio-static-online.cdn.bcebos.com/d4e7f0fe3bda4606bf907ebd12824f25376ad254d8264202bc07d955696059e4) ## 2. 项目内容 本项目基于PaddleOCR套件，以PP-OCRv3识别模型为基础，针对手写文字识别场景进行优化。 From 071d832728a8d87e25febbf29a36dde0dbac99e0 Mon Sep 17 00:00:00 2001 From: andyjpaddle Date: Wed, 3 Aug 2022 06:48:14 +0000 Subject: [PATCH 12/25] add example img --- applications/手写文字识别.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/applications/手写文字识别.md b/applications/手写文字识别.md index 4796355bb7..09d1bbab06 100644 --- a/applications/手写文字识别.md +++ b/applications/手写文字识别.md @@ -18,7 +18,7 @@ ## 1. 项目背景及意义 目前光学字符识别(OCR)技术在我们的生活当中被广泛使用，但是大多数模型在通用场景下的准确性还有待提高。针对于此我们借助飞桨提供的PaddleOCR套件较容易的实现了在垂类场景下的应用。手写体在日常生活中较为常见，然而手写体的识别却存在着很大的挑战，因为每个人的手写字体风格不一样，这对于视觉模型来说还是相当有挑战的。因此训练一个手写体识别模型具有很好的现实意义。下面给出一些手写体的示例图： -![example](https://ai-studio-static-online.cdn.bcebos.com/d4e7f0fe3bda4606bf907ebd12824f25376ad254d8264202bc07d955696059e4) +![example](https://ai-studio-static-online.cdn.bcebos.com/7a8865b2836f42d382e7c3fdaedc4d307d797fa2bcd0466e9f8b7705efff5a7b) ## 2. 项目内容 本项目基于PaddleOCR套件，以PP-OCRv3识别模型为基础，针对手写文字识别场景进行优化。 From 225a3fc4fa0d255821ea11f465ac1823e82b2aef Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Wed, 3 Aug 2022 15:31:51 +0800 Subject: [PATCH 13/25] rename layoutxlm to layoutxlm_ser --- .../{layoutxlm => layoutxlm_ser}/train_infer_python.txt | 2 +- test_tipc/docs/benchmark_train.md | 2 +- test_tipc/prepare.sh | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) rename test_tipc/configs/{layoutxlm => layoutxlm_ser}/train_infer_python.txt (98%) diff --git a/test_tipc/configs/layoutxlm/train_infer_python.txt b/test_tipc/configs/layoutxlm_ser/train_infer_python.txt similarity index 98% rename from test_tipc/configs/layoutxlm/train_infer_python.txt rename to test_tipc/configs/layoutxlm_ser/train_infer_python.txt index 32b623a8dc..887c3285ec 100644 --- a/test_tipc/configs/layoutxlm/train_infer_python.txt +++ b/test_tipc/configs/layoutxlm_ser/train_infer_python.txt @@ -1,5 +1,5 @@ ===========================train_params=========================== -model_name:layoutxlm +model_name:layoutxlm_ser python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True diff --git a/test_tipc/docs/benchmark_train.md b/test_tipc/docs/benchmark_train.md index 9076ccc40c..50cc13b92f 100644 --- a/test_tipc/docs/benchmark_train.md +++ b/test_tipc/docs/benchmark_train.md @@ -69,7 +69,7 @@ train_log/ | det_r50_vd_east_v2.0 |[config](../configs/det_r50_vd_east_v2.0/train_infer_python.txt) | 42.485 | 42.624 / 42.663 / 42.561 |0.00239083 | 67.61 |67.825/ 68.299/ 68.51| 0.00999854 | 10,000| 2,000| | det_r50_vd_pse_v2.0 |[config](../configs/det_r50_vd_pse_v2.0/train_infer_python.txt) | 16.455 | 16.517 / 16.555 / 16.353 |0.012201752 | 27.02 |27.288 / 27.152 / 27.408| 0.009340339 | 10,000| 2,000| | rec_mv3_none_bilstm_ctc_v2.0 |[config](../configs/rec_mv3_none_bilstm_ctc_v2.0/train_infer_python.txt) | 2288.358 | 2291.906 / 2293.725 / 2290.05 |0.001602197 | 2336.17 |2327.042 / 2328.093 / 2344.915| 0.007622025 | 600,000| 160,000| -| layoutxlm |[config](../configs/layoutxlm/train_infer_python.txt) | 18.001 | 18.114 / 18.107 / 18.307 |0.010924783 | 21.982 | 21.507 / 21.116 / 21.406| 0.018180127 | 1490 | 1490| +| layoutxlm_ser |[config](../configs/layoutxlm/train_infer_python.txt) | 18.001 | 18.114 / 18.107 / 18.307 |0.010924783 | 21.982 | 21.507 / 21.116 / 21.406| 0.018180127 | 1490 | 1490| | PP-Structure-table |[config](../configs/en_table_structure/train_infer_python.txt) | 14.151 | 14.077 / 14.23 / 14.25 |0.012140351 | 16.285 | 16.595 / 16.878 / 16.531 | 0.020559308 | 20,000| 5,000| | det_r50_dcn_fce_ctw_v2.0 |[config](../configs/det_r50_dcn_fce_ctw_v2.0/train_infer_python.txt) | 14.057 | 14.029 / 14.02 / 14.014 |0.001069214 | 18.298 |18.411 / 18.376 / 18.331| 0.004345228 | 10,000| 2,000| | ch_PP-OCRv3_det |[config](../configs/ch_PP-OCRv3_det/train_infer_python.txt) | 8.622 | 8.431 / 8.423 / 8.479|0.006604552 | 14.203 |14.346 14.468 14.23| 0.016450097 | 10,000| 2,000| diff --git a/test_tipc/prepare.sh b/test_tipc/prepare.sh index 35b9c73a06..709a5a6113 100644 --- a/test_tipc/prepare.sh +++ b/test_tipc/prepare.sh @@ -106,7 +106,7 @@ if [ ${MODE} = "benchmark_train" ];then ln -s ./icdar2015_benckmark ./icdar2015 cd ../ fi - if [ ${model_name} == "layoutxlm" ]; then + if [ ${model_name} == "layoutxlm_ser" ]; then pip install -r ppstructure/vqa/requirements.txt pip install paddlenlp>=2.3.5 wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/ppstructure/dataset/XFUND.tar --no-check-certificate @@ -220,7 +220,7 @@ if [ ${MODE} = "lite_train_lite_infer" ];then wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/rec_r32_gaspin_bilstm_att_train.tar --no-check-certificate cd ./pretrain_models/ && tar xf rec_r32_gaspin_bilstm_att_train.tar && cd ../ fi - if [ ${model_name} == "layoutxlm" ]; then + if [ ${model_name} == "layoutxlm_ser" ]; then pip install -r ppstructure/vqa/requirements.txt pip install paddlenlp>=2.3.5 wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/ppstructure/dataset/XFUND.tar --no-check-certificate From faa6bbef2a2928ff5757b2009520ca861b98343a Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Thu, 4 Aug 2022 10:15:12 +0800 Subject: [PATCH 14/25] rename 2.0 to 2_0 --- .../ch_ppocr_mobile_v2_0_rec/train_pact_infer_python.txt | 4 ++-- .../ch_ppocr_mobile_v2_0_rec/train_ptq_infer_python.txt | 2 +- test_tipc/test_paddle2onnx.sh | 6 +++--- test_tipc/test_serving_infer_python.sh | 6 +++--- 4 files changed, 9 insertions(+), 9 deletions(-) diff --git a/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_pact_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_pact_infer_python.txt index 631518ffc4..9c1223f41a 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_pact_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_pact_infer_python.txt @@ -14,7 +14,7 @@ null:null ## trainer:pact_train norm_train:null -pact_train:deploy/slim/quantization/quant.py -c test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/rec_chinese_lite_train_v2.0.yml -o +pact_train:deploy/slim/quantization/quant.py -c test_tipc/configs/ch_ppocr_mobile_v2_0_rec_PACT/rec_chinese_lite_train_v2.0.yml -o fpgm_train:null distill_train:null null:null @@ -28,7 +28,7 @@ null:null Global.save_inference_dir:./output/ Global.checkpoints: norm_export:null -quant_export:deploy/slim/quantization/export_model.py -c test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/rec_chinese_lite_train_v2.0.yml -o +quant_export:deploy/slim/quantization/export_model.py -c test_tipc/configs/ch_ppocr_mobile_v2_0_rec_PACT/rec_chinese_lite_train_v2.0.yml -o fpgm_export:null distill_export:null export1:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_ptq_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_ptq_infer_python.txt index cd0828f3c6..df47f328bd 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_ptq_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2_0_rec/train_ptq_infer_python.txt @@ -4,7 +4,7 @@ python:python3.7 Global.pretrained_model:null Global.save_inference_dir:null infer_model:./inference/ch_ppocr_mobile_v2.0_rec_infer/ -infer_export:deploy/slim/quantization/quant_kl.py -c test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/rec_chinese_lite_train_v2.0.yml -o +infer_export:deploy/slim/quantization/quant_kl.py -c test_tipc/configs/ch_ppocr_mobile_v2_0_rec_KL/rec_chinese_lite_train_v2.0.yml -o infer_quant:True inference:tools/infer/predict_rec.py --rec_image_shape="3,32,320" --use_gpu:False|True diff --git a/test_tipc/test_paddle2onnx.sh b/test_tipc/test_paddle2onnx.sh index 356bc98041..78d79d0b8e 100644 --- a/test_tipc/test_paddle2onnx.sh +++ b/test_tipc/test_paddle2onnx.sh @@ -54,7 +54,7 @@ function func_paddle2onnx(){ _script=$1 # paddle2onnx - if [ ${model_name} = "ch_PP-OCRv2" ] || [ ${model_name} = "ch_PP-OCRv3" ] || [ ${model_name} = "ch_ppocr_mobile_v2.0" ] || [ ${model_name} = "ch_ppocr_server_v2.0" ]; then + if [ ${model_name} = "ch_PP-OCRv2" ] || [ ${model_name} = "ch_PP-OCRv3" ] || [ ${model_name} = "ch_ppocr_mobile_v2_0" ] || [ ${model_name} = "ch_ppocr_server_v2_0" ]; then # trans det set_dirname=$(func_set_params "--model_dir" "${det_infer_model_dir_value}") set_model_filename=$(func_set_params "${model_filename_key}" "${model_filename_value}") @@ -113,7 +113,7 @@ function func_paddle2onnx(){ _save_log_path="${LOG_PATH}/paddle2onnx_infer_cpu.log" set_gpu=$(func_set_params "${use_gpu_key}" "${use_gpu}") set_img_dir=$(func_set_params "${image_dir_key}" "${image_dir_value}") - if [ ${model_name} = "ch_PP-OCRv2" ] || [ ${model_name} = "ch_PP-OCRv3" ] || [ ${model_name} = "ch_ppocr_mobile_v2.0" ] || [ ${model_name} = "ch_ppocr_server_v2.0" ]; then + if [ ${model_name} = "ch_PP-OCRv2" ] || [ ${model_name} = "ch_PP-OCRv3" ] || [ ${model_name} = "ch_ppocr_mobile_v2_0" ] || [ ${model_name} = "ch_ppocr_server_v2_0" ]; then set_det_model_dir=$(func_set_params "${det_model_key}" "${det_save_file_value}") set_rec_model_dir=$(func_set_params "${rec_model_key}" "${rec_save_file_value}") infer_model_cmd="${python} ${inference_py} ${set_gpu} ${set_img_dir} ${set_det_model_dir} ${set_rec_model_dir} --use_onnx=True > ${_save_log_path} 2>&1 " @@ -132,7 +132,7 @@ function func_paddle2onnx(){ _save_log_path="${LOG_PATH}/paddle2onnx_infer_gpu.log" set_gpu=$(func_set_params "${use_gpu_key}" "${use_gpu}") set_img_dir=$(func_set_params "${image_dir_key}" "${image_dir_value}") - if [ ${model_name} = "ch_PP-OCRv2" ] || [ ${model_name} = "ch_PP-OCRv3" ] || [ ${model_name} = "ch_ppocr_mobile_v2.0" ] || [ ${model_name} = "ch_ppocr_server_v2.0" ]; then + if [ ${model_name} = "ch_PP-OCRv2" ] || [ ${model_name} = "ch_PP-OCRv3" ] || [ ${model_name} = "ch_ppocr_mobile_v2_0" ] || [ ${model_name} = "ch_ppocr_server_v2_0" ]; then set_det_model_dir=$(func_set_params "${det_model_key}" "${det_save_file_value}") set_rec_model_dir=$(func_set_params "${rec_model_key}" "${rec_save_file_value}") infer_model_cmd="${python} ${inference_py} ${set_gpu} ${set_img_dir} ${set_det_model_dir} ${set_rec_model_dir} --use_onnx=True > ${_save_log_path} 2>&1 " diff --git a/test_tipc/test_serving_infer_python.sh b/test_tipc/test_serving_infer_python.sh index 4ccccc06e2..4b7dfcf785 100644 --- a/test_tipc/test_serving_infer_python.sh +++ b/test_tipc/test_serving_infer_python.sh @@ -71,7 +71,7 @@ function func_serving(){ # pdserving set_model_filename=$(func_set_params "${model_filename_key}" "${model_filename_value}") set_params_filename=$(func_set_params "${params_filename_key}" "${params_filename_value}") - if [ ${model_name} = "ch_PP-OCRv2" ] || [ ${model_name} = "ch_PP-OCRv3" ] || [ ${model_name} = "ch_ppocr_mobile_v2.0" ] || [ ${model_name} = "ch_ppocr_server_v2.0" ]; then + if [ ${model_name} = "ch_PP-OCRv2" ] || [ ${model_name} = "ch_PP-OCRv3" ] || [ ${model_name} = "ch_ppocr_mobile_v2_0" ] || [ ${model_name} = "ch_ppocr_server_v2_0" ]; then # trans det set_dirname=$(func_set_params "--dirname" "${det_infer_model_dir_value}") set_serving_server=$(func_set_params "--serving_server" "${det_serving_server_value}") @@ -120,7 +120,7 @@ function func_serving(){ for threads in ${web_cpu_threads_list[*]}; do set_cpu_threads=$(func_set_params "${web_cpu_threads_key}" "${threads}") server_log_path="${LOG_PATH}/python_server_cpu_usemkldnn_${use_mkldnn}_threads_${threads}.log" - if [ ${model_name} = "ch_PP-OCRv2" ] || [ ${model_name} = "ch_PP-OCRv3" ] || [ ${model_name} = "ch_ppocr_mobile_v2.0" ] || [ ${model_name} = "ch_ppocr_server_v2.0" ]; then + if [ ${model_name} = "ch_PP-OCRv2" ] || [ ${model_name} = "ch_PP-OCRv3" ] || [ ${model_name} = "ch_ppocr_mobile_v2_0" ] || [ ${model_name} = "ch_ppocr_server_v2_0" ]; then set_det_model_config=$(func_set_params "${det_server_key}" "${det_server_value}") set_rec_model_config=$(func_set_params "${rec_server_key}" "${rec_server_value}") web_service_cmd="nohup ${python} ${web_service_py} ${web_use_gpu_key}="" ${web_use_mkldnn_key}=${use_mkldnn} ${set_cpu_threads} ${set_det_model_config} ${set_rec_model_config} > ${server_log_path} 2>&1 &" @@ -171,7 +171,7 @@ function func_serving(){ device_type=2 fi set_precision=$(func_set_params "${web_precision_key}" "${precision}") - if [ ${model_name} = "ch_PP-OCRv2" ] || [ ${model_name} = "ch_PP-OCRv3" ] || [ ${model_name} = "ch_ppocr_mobile_v2.0" ] || [ ${model_name} = "ch_ppocr_server_v2.0" ]; then + if [ ${model_name} = "ch_PP-OCRv2" ] || [ ${model_name} = "ch_PP-OCRv3" ] || [ ${model_name} = "ch_ppocr_mobile_v2_0" ] || [ ${model_name} = "ch_ppocr_server_v2_0" ]; then set_det_model_config=$(func_set_params "${det_server_key}" "${det_server_value}") set_rec_model_config=$(func_set_params "${rec_server_key}" "${rec_server_value}") web_service_cmd="nohup ${python} ${web_service_py} ${set_tensorrt} ${set_precision} ${set_det_model_config} ${set_rec_model_config} > ${server_log_path} 2>&1 &" From fbcb5b6f1768641e2021e8f230bf312cfe74eb53 Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Thu, 4 Aug 2022 10:20:13 +0800 Subject: [PATCH 15/25] rename 2.0 to 2_0 --- test_tipc/prepare_lite_cpp.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/test_tipc/prepare_lite_cpp.sh b/test_tipc/prepare_lite_cpp.sh index 9148cb5dd7..0d3a5ca45a 100644 --- a/test_tipc/prepare_lite_cpp.sh +++ b/test_tipc/prepare_lite_cpp.sh @@ -49,7 +49,7 @@ model_path=./inference_models for model in ${lite_model_list[*]}; do if [[ $model =~ "PP-OCRv2" ]]; then inference_model_url=https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/${model}.tar - elif [[ $model =~ "v2.0" ]]; then + elif [[ $model =~ "v2_0" ]]; then inference_model_url=https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/${model}.tar elif [[ $model =~ "PP-OCRv3" ]]; then inference_model_url=https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/${model}.tar From 604e59339e219b16b0550b459c315f843abaadcd Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Thu, 4 Aug 2022 12:14:31 +0800 Subject: [PATCH 16/25] add paddlenlp install --- test_tipc/prepare.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/test_tipc/prepare.sh b/test_tipc/prepare.sh index 709a5a6113..931024382e 100644 --- a/test_tipc/prepare.sh +++ b/test_tipc/prepare.sh @@ -108,7 +108,7 @@ if [ ${MODE} = "benchmark_train" ];then fi if [ ${model_name} == "layoutxlm_ser" ]; then pip install -r ppstructure/vqa/requirements.txt - pip install paddlenlp>=2.3.5 + pip install paddlenlp\>=2.3.5 --force-reinstall wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/ppstructure/dataset/XFUND.tar --no-check-certificate cd ./train_data/ && tar xf XFUND.tar # expand gt.txt 10 times @@ -222,7 +222,7 @@ if [ ${MODE} = "lite_train_lite_infer" ];then fi if [ ${model_name} == "layoutxlm_ser" ]; then pip install -r ppstructure/vqa/requirements.txt - pip install paddlenlp>=2.3.5 + pip install paddlenlp\>=2.3.5 --force-reinstall wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/ppstructure/dataset/XFUND.tar --no-check-certificate cd ./train_data/ && tar xf XFUND.tar cd ../ From d364baa75f686872d513c084a665e2eb83764cca Mon Sep 17 00:00:00 2001 From: Evezerest <50011306+Evezerest@users.noreply.github.com> Date: Thu, 4 Aug 2022 13:20:48 +0800 Subject: [PATCH 17/25] Update README_ch.md --- PPOCRLabel/README_ch.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/PPOCRLabel/README_ch.md b/PPOCRLabel/README_ch.md index 800ef14004..107f902a68 100644 --- a/PPOCRLabel/README_ch.md +++ b/PPOCRLabel/README_ch.md @@ -131,7 +131,7 @@ pip3 install dist/PPOCRLabel-1.0.2-py2.py3-none-any.whl -i https://mirror.baidu. > 注意：如果表格中存在空白单元格，同样需要使用一个标注框将其标出，使得单元格总数与图像中保持一致。 -3. **调整单元格顺序：**点击软件`视图-显示框编号` 打开标注框序号，在软件界面右侧拖动 `识别结果` 一栏下的所有结果，使得标注框编号按照从左到右，从上到下的顺序排列 +3. **调整单元格顺序**：点击软件`视图-显示框编号` 打开标注框序号，在软件界面右侧拖动 `识别结果` 一栏下的所有结果，使得标注框编号按照从左到右，从上到下的顺序排列，按行依次标注。 4. 标注表格结构：**在外部Excel软件中，将存在文字的单元格标记为任意标识符（如 `1` ）**，保证Excel中的单元格合并情况与原图相同即可（即不需要Excel中的单元格文字与图片中的文字完全相同） From 12aa058cb135df6030d6c0ffa99318fc24f06f0b Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Fri, 5 Aug 2022 11:04:14 +0800 Subject: [PATCH 18/25] reduce layoutxml batchsize in litetrainliteinfer model --- test_tipc/configs/layoutxlm_ser/train_infer_python.txt | 2 +- test_tipc/prepare.sh | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/test_tipc/configs/layoutxlm_ser/train_infer_python.txt b/test_tipc/configs/layoutxlm_ser/train_infer_python.txt index 887c3285ec..6d05d413e1 100644 --- a/test_tipc/configs/layoutxlm_ser/train_infer_python.txt +++ b/test_tipc/configs/layoutxlm_ser/train_infer_python.txt @@ -6,7 +6,7 @@ Global.use_gpu:True|True Global.auto_cast:fp32 Global.epoch_num:lite_train_lite_infer=1|whole_train_whole_infer=17 Global.save_model_dir:./output/ -Train.loader.batch_size_per_card:lite_train_lite_infer=8|whole_train_whole_infer=8 +Train.loader.batch_size_per_card:lite_train_lite_infer=4|whole_train_whole_infer=8 Architecture.Backbone.checkpoints:null train_model_name:latest train_infer_img_dir:ppstructure/docs/vqa/input/zh_val_42.jpg diff --git a/test_tipc/prepare.sh b/test_tipc/prepare.sh index 931024382e..76543f39e4 100644 --- a/test_tipc/prepare.sh +++ b/test_tipc/prepare.sh @@ -108,7 +108,7 @@ if [ ${MODE} = "benchmark_train" ];then fi if [ ${model_name} == "layoutxlm_ser" ]; then pip install -r ppstructure/vqa/requirements.txt - pip install paddlenlp\>=2.3.5 --force-reinstall + pip install paddlenlp\>=2.3.5 --force-reinstall -i https://mirrors.aliyun.com/pypi/simple/ wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/ppstructure/dataset/XFUND.tar --no-check-certificate cd ./train_data/ && tar xf XFUND.tar # expand gt.txt 10 times @@ -222,7 +222,7 @@ if [ ${MODE} = "lite_train_lite_infer" ];then fi if [ ${model_name} == "layoutxlm_ser" ]; then pip install -r ppstructure/vqa/requirements.txt - pip install paddlenlp\>=2.3.5 --force-reinstall + pip install paddlenlp\>=2.3.5 --force-reinstall -i https://mirrors.aliyun.com/pypi/simple/ wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/ppstructure/dataset/XFUND.tar --no-check-certificate cd ./train_data/ && tar xf XFUND.tar cd ../ From d0efcc74c92f392b28f81ede3d79b0a86de075f2 Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Fri, 5 Aug 2022 12:41:03 +0800 Subject: [PATCH 19/25] convert fp16 params to fp32 when params is fp16 format --- ppocr/utils/save_load.py | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/ppocr/utils/save_load.py b/ppocr/utils/save_load.py index 3647111fdd..5d43d656f2 100644 --- a/ppocr/utils/save_load.py +++ b/ppocr/utils/save_load.py @@ -90,12 +90,16 @@ def load_model(config, model, optimizer=None, model_type='det'): params = paddle.load(checkpoints + '.pdparams') state_dict = model.state_dict() new_state_dict = {} + is_float16 = False for key, value in state_dict.items(): if key not in params: logger.warning("{} not in loaded params {} !".format( key, params.keys())) continue pre_value = params[key] + if pre_value.dtype == paddle.float16: + pre_value = pre_value.astype(paddle.float32) + is_float16 = True if list(value.shape) == list(pre_value.shape): new_state_dict[key] = pre_value else: @@ -104,6 +108,10 @@ def load_model(config, model, optimizer=None, model_type='det'): format(key, value.shape, pre_value.shape)) model.set_state_dict(new_state_dict) + if is_float16: + logger.info( + "The parameter type is float16, which is converted to float32 when loading" + ) if optimizer is not None: if os.path.exists(checkpoints + '.pdopt'): optim_dict = paddle.load(checkpoints + '.pdopt') @@ -138,17 +146,26 @@ def load_pretrained_params(model, path): params = paddle.load(path + '.pdparams') state_dict = model.state_dict() new_state_dict = {} + is_float16 = False for k1 in params.keys(): if k1 not in state_dict.keys(): logger.warning("The pretrained params {} not in model".format(k1)) else: + if params[k1].dtype == paddle.float16: + params[k1] = params[k1].astype(paddle.float32) + is_float16 = True if list(state_dict[k1].shape) == list(params[k1].shape): new_state_dict[k1] = params[k1] else: logger.warning( "The shape of model params {} {} not matched with loaded params {} {} !". format(k1, state_dict[k1].shape, k1, params[k1].shape)) + model.set_state_dict(new_state_dict) + if is_float16: + logger.info( + "The parameter type is float16, which is converted to float32 when loading" + ) logger.info("load pretrain successful from {}".format(path)) return model From c6738f4c5386517bef843f04d3d9fe9aa9714bdc Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Fri, 5 Aug 2022 12:50:25 +0800 Subject: [PATCH 20/25] convert fp16 params to fp32 when params is fp16 format --- ppocr/utils/save_load.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/ppocr/utils/save_load.py b/ppocr/utils/save_load.py index 5d43d656f2..53bf5fa9d0 100644 --- a/ppocr/utils/save_load.py +++ b/ppocr/utils/save_load.py @@ -53,6 +53,7 @@ def load_model(config, model, optimizer=None, model_type='det'): checkpoints = global_config.get('checkpoints') pretrained_model = global_config.get('pretrained_model') best_model_dict = {} + is_float16 = False if model_type == 'vqa': checkpoints = config['Architecture']['Backbone']['checkpoints'] @@ -90,7 +91,6 @@ def load_model(config, model, optimizer=None, model_type='det'): params = paddle.load(checkpoints + '.pdparams') state_dict = model.state_dict() new_state_dict = {} - is_float16 = False for key, value in state_dict.items(): if key not in params: logger.warning("{} not in loaded params {} !".format( @@ -107,7 +107,6 @@ def load_model(config, model, optimizer=None, model_type='det'): "The shape of model params {} {} not matched with loaded params shape {} !". format(key, value.shape, pre_value.shape)) model.set_state_dict(new_state_dict) - if is_float16: logger.info( "The parameter type is float16, which is converted to float32 when loading" @@ -130,9 +129,10 @@ def load_model(config, model, optimizer=None, model_type='det'): best_model_dict['start_epoch'] = states_dict['epoch'] + 1 logger.info("resume from {}".format(checkpoints)) elif pretrained_model: - load_pretrained_params(model, pretrained_model) + is_float16 = load_pretrained_params(model, pretrained_model) else: logger.info('train from scratch') + best_model_dict['is_float16'] = is_float16 return best_model_dict @@ -167,7 +167,7 @@ def load_pretrained_params(model, path): "The parameter type is float16, which is converted to float32 when loading" ) logger.info("load pretrain successful from {}".format(path)) - return model + return is_float16 def save_model(model, From 8c4e1dccf2e11dbb4f50eef3f9a325ba31076152 Mon Sep 17 00:00:00 2001 From: MissPenguin Date: Sat, 6 Aug 2022 13:16:51 +0800 Subject: [PATCH 21/25] refine doc (#7110) --- applications/README.md | 99 ++++++++++++++++++-------- applications/高精度中文识别模型.md | 2 +- doc/doc_ch/PP-OCRv3_introduction.md | 5 +- doc/doc_en/PP-OCRv3_introduction_en.md | 5 +- 4 files changed, 75 insertions(+), 36 deletions(-) diff --git a/applications/README.md b/applications/README.md index eba1e205dc..017c2a9f6f 100644 --- a/applications/README.md +++ b/applications/README.md @@ -1,41 +1,78 @@ +[English](README_en.md) | 简体中文 + # 场景应用 PaddleOCR场景应用覆盖通用，制造、金融、交通行业的主要OCR垂类应用，在PP-OCR、PP-Structure的通用能力基础之上，以notebook的形式展示利用场景数据微调、模型优化方法、数据增广等内容，为开发者快速落地OCR应用提供示范与启发。 -> 如需下载全部垂类模型，可以扫描下方二维码，关注公众号填写问卷后，加入PaddleOCR官方交流群获取20G OCR学习大礼包（内含《动手学OCR》电子书、课程回放视频、前沿论文等重磅资料） +- [教程文档](#1) + - [通用](#11) + - [制造](#12) + - [金融](#13) + - [交通](#14) + +- [模型下载](#2) + + + +## 教程文档 + + + +### 通用 + +| 类别 | 亮点 | 模型下载 | 教程 | +| ---------------------- | ------------ | -------------- | --------------------------------------- | +| 高精度中文识别模型SVTR | 比PP-OCRv3识别模型精度高3%，可用于数据挖掘或对预测效率要求不高的场景。| [模型下载](#2) | [中文](./高精度中文识别模型.md)/English | +| 手写体识别 | 新增字形支持 | | | + + + +### 制造 + +| 类别 | 亮点 | 模型下载 | 教程 | 示例图 | +| -------------- | ------------------------------ | -------------- | ------------------------------------------------------------ | ------------------------------------------------------------ | +| 数码管识别 | 数码管数据合成、漏识别调优 | [模型下载](#2) | [中文](./光功率计数码管字符识别/光功率计数码管字符识别.md)/English | | +| 液晶屏读数识别 | 检测模型蒸馏、Serving部署 | [模型下载](#2) | [中文](./液晶屏读数识别.md)/English | | +| 包装生产日期 | 点阵字符合成、过曝过暗文字识别 | [模型下载](#2) | [中文](./包装生产日期识别.md)/English | | +| PCB文字识别 | 小尺寸文本检测与识别 | [模型下载](#2) | [中文](./PCB字符识别/PCB字符识别.md)/English | | +| 电表识别 | 大分辨率图像检测调优 | [模型下载](#2) | | | +| 液晶屏缺陷检测 | 非文字字符识别 | | | | + + + +### 金融 + +| 类别 | 亮点 | 模型下载 | 教程 | 示例图 | +| -------------- | ------------------------ | -------------- | ----------------------------------- | ------------------------------------------------------------ | +| 表单VQA | 多模态通用表单结构化提取 | [模型下载](#2) | [中文](./多模态表单识别.md)/English | | +| 增值税发票 | 尽请期待 | | | | +| 印章检测与识别 | 端到端弯曲文本识别 | | | | +| 通用卡证识别 | 通用结构化提取 | | | | +| 身份证识别 | 结构化提取、图像阴影 | | | | +| 合同比对 | 密集文本检测、NLP串联 | | | | + + + +### 交通 + +| 类别 | 亮点 | 模型下载 | 教程 | 示例图 | +| ----------------- | ------------------------------ | -------------- | ----------------------------------- | ------------------------------------------------------------ | +| 车牌识别 | 多角度图像、轻量模型、端侧部署 | [模型下载](#2) | [中文](./轻量级车牌识别.md)/English | | +| 驾驶证/行驶证识别 | 尽请期待 | | | | +| 快递单识别 | 尽请期待 | | | | + + + +## 模型下载 + +如需下载上述场景中已经训练好的垂类模型，可以扫描下方二维码，关注公众号填写问卷后，加入PaddleOCR官方交流群获取20G OCR学习大礼包（内含《动手学OCR》电子书、课程回放视频、前沿论文等重磅资料）
+如果您是企业开发者且未在上述场景中找到合适的方案，可以填写[OCR应用合作调研问卷](https://paddle.wjx.cn/vj/QwF7GKw.aspx)，免费与官方团队展开不同层次的合作，包括但不限于问题抽象、确定技术方案、项目答疑、共同研发等。如果您已经使用PaddleOCR落地项目，也可以填写此问卷，与飞桨平台共同宣传推广，提升企业技术品宣。期待您的提交！ -> 如果您是企业开发者且未在下述场景中找到合适的方案，可以填写[OCR应用合作调研问卷](https://paddle.wjx.cn/vj/QwF7GKw.aspx)，免费与官方团队展开不同层次的合作，包括但不限于问题抽象、确定技术方案、项目答疑、共同研发等。如果您已经使用PaddleOCR落地项目，也可以填写此问卷，与飞桨平台共同宣传推广，提升企业技术品宣。期待您的提交！ - -## 通用 - -| 类别 | 亮点 | 类别 | 亮点 | -| ------------------------------------------------- | -------- | ---------- | ------------ | -| [高精度中文识别模型SVTR](./高精度中文识别模型.md) | 新增模型 | 手写体识别 | 新增字形支持 | - -## 制造 - -| 类别 | 亮点 | 类别 | 亮点 | -| ------------------------------------------------------------ | ------------------------------ | ------------------------------------------- | -------------------- | -| [数码管识别](./光功率计数码管字符识别/光功率计数码管字符识别.md) | 数码管数据合成、漏识别调优 | 电表识别 | 大分辨率图像检测调优 | -| [液晶屏读数识别](./液晶屏读数识别.md) | 检测模型蒸馏、Serving部署 | [PCB文字识别](./PCB字符识别/PCB字符识别.md) | 小尺寸文本检测与识别 | -| [包装生产日期](./包装生产日期识别.md) | 点阵字符合成、过曝过暗文字识别 | 液晶屏缺陷检测 | 非文字字符识别 | - -## 金融 - -| 类别 | 亮点 | 类别 | 亮点 | -| ------------------------------ | ------------------------ | ------------ | --------------------- | -| [表单VQA](./多模态表单识别.md) | 多模态通用表单结构化提取 | 通用卡证识别 | 通用结构化提取 | -| 增值税发票 | 尽请期待 | 身份证识别 | 结构化提取、图像阴影 | -| 印章检测与识别 | 端到端弯曲文本识别 | 合同比对 | 密集文本检测、NLP串联 | - -## 交通 - -| 类别 | 亮点 | 类别 | 亮点 | -| ------------------------------- | ------------------------------ | ---------- | -------- | -| [车牌识别](./轻量级车牌识别.md) | 多角度图像、轻量模型、端侧部署 | 快递单识别 | 尽请期待 | -| 驾驶证/行驶证识别 | 尽请期待 | | | \ No newline at end of file + +traffic + diff --git a/applications/高精度中文识别模型.md b/applications/高精度中文识别模型.md index 3c31af42ee..4e71e23300 100644 --- a/applications/高精度中文识别模型.md +++ b/applications/高精度中文识别模型.md @@ -2,7 +2,7 @@ ## 1. 简介 -PP-OCRv3是百度开源的超轻量级场景文本检测识别模型库，其中超轻量的场景中文识别模型SVTR_LCNet使用了SVTR算法结构。为了保证速度，SVTR_LCNet将SVTR模型的Local Blocks替换为LCNet，使用两层Global Blocks。在中文场景中，PP-OCRv3识别主要使用如下优化策略： +PP-OCRv3是百度开源的超轻量级场景文本检测识别模型库，其中超轻量的场景中文识别模型SVTR_LCNet使用了SVTR算法结构。为了保证速度，SVTR_LCNet将SVTR模型的Local Blocks替换为LCNet，使用两层Global Blocks。在中文场景中，PP-OCRv3识别主要使用如下优化策略（[详细技术报告](../doc/doc_ch/PP-OCRv3_introduction.md)）： - GTC：Attention指导CTC训练策略； - TextConAug：挖掘文字上下文信息的数据增广策略； - TextRotNet：自监督的预训练模型； diff --git a/doc/doc_ch/PP-OCRv3_introduction.md b/doc/doc_ch/PP-OCRv3_introduction.md index 3c17921a24..ddeb78d74f 100644 --- a/doc/doc_ch/PP-OCRv3_introduction.md +++ b/doc/doc_ch/PP-OCRv3_introduction.md @@ -53,10 +53,11 @@ PP-OCRv3检测模型是对PP-OCRv2中的[CML](https://arxiv.org/pdf/2109.03144.p |序号|策略|模型大小|hmean|速度（cpu + mkldnn)| |-|-|-|-|-| -|baseline teacher|DB-R50|99M|83.5%|260ms| +|baseline teacher|PP-OCR server|49M|83.2%|171ms| |teacher1|DB-R50-LK-PAN|124M|85.0%|396ms| |teacher2|DB-R50-LK-PAN-DML|124M|86.0%|396ms| |baseline student|PP-OCRv2|3M|83.2%|117ms| +|student0|DB-MV3-RSE-FPN|3.6M|84.5%|124ms| |student1|DB-MV3-CML（teacher2）|3M|84.3%|117ms| |student2|DB-MV3-RSE-FPN-CML（teacher2）|3.6M|85.4%|124ms| @@ -184,7 +185,7 @@ UDML（Unified-Deep Mutual Learning）联合互学习是PP-OCRv2中就采用的 **（6）UIM：无标注数据挖掘方案** -UIM（Unlabeled Images Mining）是一种非常简单的无标注数据挖掘方案。核心思想是利用高精度的文本识别大模型对无标注数据进行预测，获取伪标签，并且选择预测置信度高的样本作为训练数据，用于训练小模型。使用该策略，识别模型的准确率进一步提升到79.4%（+1%）。 +UIM（Unlabeled Images Mining）是一种非常简单的无标注数据挖掘方案。核心思想是利用高精度的文本识别大模型对无标注数据进行预测，获取伪标签，并且选择预测置信度高的样本作为训练数据，用于训练小模型。使用该策略，识别模型的准确率进一步提升到79.4%（+1%）。实际操作中，我们使用全量数据集训练高精度SVTR-Tiny模型（acc=82.5%）进行数据挖掘，点击获取[模型下载地址和使用教程](../../applications/高精度中文识别模型.md)。
diff --git a/doc/doc_en/PP-OCRv3_introduction_en.md b/doc/doc_en/PP-OCRv3_introduction_en.md index 481e0b8174..815ad9b0e5 100644 --- a/doc/doc_en/PP-OCRv3_introduction_en.md +++ b/doc/doc_en/PP-OCRv3_introduction_en.md @@ -55,10 +55,11 @@ The ablation experiments are as follows: |ID|Strategy|Model Size|Hmean|The Inference Time（cpu + mkldnn)| |-|-|-|-|-| -|baseline teacher|DB-R50|99M|83.5%|260ms| +|baseline teacher|PP-OCR server|49M|83.2%|171ms| |teacher1|DB-R50-LK-PAN|124M|85.0%|396ms| |teacher2|DB-R50-LK-PAN-DML|124M|86.0%|396ms| |baseline student|PP-OCRv2|3M|83.2%|117ms| +|student0|DB-MV3-RSE-FPN|3.6M|84.5%|124ms| |student1|DB-MV3-CML（teacher2）|3M|84.3%|117ms| |student2|DB-MV3-RSE-FPN-CML（teacher2）|3.6M|85.4%|124ms| @@ -199,7 +200,7 @@ UDML (Unified-Deep Mutual Learning) is a strategy proposed in PP-OCRv2 which is **（6）UIM：Unlabeled Images Mining** -UIM (Unlabeled Images Mining) is a very simple unlabeled data mining strategy. The main idea is to use a high-precision text recognition model to predict unlabeled images to obtain pseudo-labels, and select samples with high prediction confidence as training data for training lightweight models. Using this strategy, the accuracy of the recognition model is further improved to 79.4% (+1%). +UIM (Unlabeled Images Mining) is a very simple unlabeled data mining strategy. The main idea is to use a high-precision text recognition model to predict unlabeled images to obtain pseudo-labels, and select samples with high prediction confidence as training data for training lightweight models. Using this strategy, the accuracy of the recognition model is further improved to 79.4% (+1%). In practice, we use the full data set to train the high-precision SVTR_Tiny model (acc=82.5%) for data mining. [SVTR_Tiny model download and tutorial](../../applications/高精度中文识别模型.md).
From 9e4ae9dc122dda2c77ca55e63a520231ae26bebd Mon Sep 17 00:00:00 2001 From: littletomatodonkey Date: Sat, 6 Aug 2022 15:41:20 +0800 Subject: [PATCH 22/25] add vqa code (#7096) * add vqa code * add order ocr info * rename tb-yx order * polish configs * add trt offline-tuning * fix seed and remove unused configs --- .../re_layoutlmv2_xfund_zh.yml} | 4 +- .../re_layoutxlm_xfund_zh.yml} | 8 +- .../ser_layoutlm_xfund_zh.yml} | 5 +- .../ser_layoutlmv2_xfund_zh.yml} | 1 + .../ser_layoutxlm_xfund_zh.yml} | 1 + configs/kie/{ => sdmgr}/kie_unet_sdmgr.yml | 0 .../re_vi_layoutxlm_xfund_zh.yml} | 40 ++-- .../re_vi_layoutxlm_xfund_zh_udml.yml | 175 +++++++++++++++++ .../ser_vi_layoutxlm_xfund_zh.yml} | 40 ++-- .../ser_vi_layoutxlm_xfund_zh_udml.yml | 183 ++++++++++++++++++ configs/vqa/re/layoutlmv2_funsd.yml | 125 ------------ configs/vqa/ser/layoutlm_sroie.yml | 124 ------------ configs/vqa/ser/layoutlmv2_funsd.yml | 123 ------------ configs/vqa/ser/layoutlmv2_sroie.yml | 123 ------------ configs/vqa/ser/layoutxlm_funsd.yml | 123 ------------ configs/vqa/ser/layoutxlm_sroie.yml | 123 ------------ configs/vqa/ser/layoutxlm_wildreceipt.yml | 123 ------------ ppocr/data/imaug/label_ops.py | 20 +- ppocr/data/imaug/vqa/__init__.py | 2 - ppocr/data/imaug/vqa/augment.py | 30 ++- ppocr/losses/basic_loss.py | 19 +- ppocr/losses/combined_loss.py | 3 + ppocr/losses/distillation_loss.py | 132 +++++++++++++ ppocr/losses/vqa_token_layoutlm_loss.py | 14 +- ppocr/metrics/distillation_metric.py | 2 + ppocr/modeling/architectures/base_model.py | 38 ++-- ppocr/modeling/backbones/vqa_layoutlm.py | 127 ++++++++---- ppocr/postprocess/__init__.py | 37 +++- .../vqa_token_re_layoutlm_postprocess.py | 22 +++ .../vqa_token_ser_layoutlm_postprocess.py | 22 +++ ppocr/utils/save_load.py | 16 +- tools/infer/utility.py | 14 +- tools/infer_vqa_token_ser_re.py | 30 ++- 33 files changed, 835 insertions(+), 1014 deletions(-) rename configs/{vqa/re/layoutlmv2_xund_zh.yml => kie/layoutlm_series/re_layoutlmv2_xfund_zh.yml} (98%) rename configs/{vqa/re/layoutxlm_xfund_zh.yml => kie/layoutlm_series/re_layoutxlm_xfund_zh.yml} (95%) rename configs/{vqa/ser/layoutlm_xfund_zh.yml => kie/layoutlm_series/ser_layoutlm_xfund_zh.yml} (96%) rename configs/{vqa/ser/layoutlmv2_xfund_zh.yml => kie/layoutlm_series/ser_layoutlmv2_xfund_zh.yml} (99%) rename configs/{vqa/ser/layoutxlm_xfund_zh.yml => kie/layoutlm_series/ser_layoutxlm_xfund_zh.yml} (99%) rename configs/kie/{ => sdmgr}/kie_unet_sdmgr.yml (100%) rename configs/{vqa/re/layoutxlm_funsd.yml => kie/vi_layoutxlm/re_vi_layoutxlm_xfund_zh.yml} (73%) create mode 100644 configs/kie/vi_layoutxlm/re_vi_layoutxlm_xfund_zh_udml.yml rename configs/{vqa/ser/layoutlm_funsd.yml => kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh.yml} (72%) create mode 100644 configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh_udml.yml delete mode 100644 configs/vqa/re/layoutlmv2_funsd.yml delete mode 100644 configs/vqa/ser/layoutlm_sroie.yml delete mode 100644 configs/vqa/ser/layoutlmv2_funsd.yml delete mode 100644 configs/vqa/ser/layoutlmv2_sroie.yml delete mode 100644 configs/vqa/ser/layoutxlm_funsd.yml delete mode 100644 configs/vqa/ser/layoutxlm_sroie.yml delete mode 100644 configs/vqa/ser/layoutxlm_wildreceipt.yml diff --git a/configs/vqa/re/layoutlmv2_xund_zh.yml b/configs/kie/layoutlm_series/re_layoutlmv2_xfund_zh.yml similarity index 98% rename from configs/vqa/re/layoutlmv2_xund_zh.yml rename to configs/kie/layoutlm_series/re_layoutlmv2_xfund_zh.yml index 986b9b5cef..4b330d8d58 100644 --- a/configs/vqa/re/layoutlmv2_xund_zh.yml +++ b/configs/kie/layoutlm_series/re_layoutlmv2_xfund_zh.yml @@ -6,11 +6,11 @@ Global: save_model_dir: ./output/re_layoutlmv2_xfund_zh save_epoch_step: 2000 # evaluation is run every 10 iterations after the 0th iteration - eval_batch_step: [ 0, 57 ] + eval_batch_step: [ 0, 19 ] cal_metric_during_train: False save_inference_dir: use_visualdl: False - seed: 2048 + seed: 2022 infer_img: ppstructure/docs/vqa/input/zh_val_21.jpg save_res_path: ./output/re_layoutlmv2_xfund_zh/res/ diff --git a/configs/vqa/re/layoutxlm_xfund_zh.yml b/configs/kie/layoutlm_series/re_layoutxlm_xfund_zh.yml similarity index 95% rename from configs/vqa/re/layoutxlm_xfund_zh.yml rename to configs/kie/layoutlm_series/re_layoutxlm_xfund_zh.yml index d8585bb725..a092106eea 100644 --- a/configs/vqa/re/layoutxlm_xfund_zh.yml +++ b/configs/kie/layoutlm_series/re_layoutxlm_xfund_zh.yml @@ -1,9 +1,9 @@ Global: use_gpu: True - epoch_num: &epoch_num 200 + epoch_num: &epoch_num 130 log_smooth_window: 10 print_batch_step: 10 - save_model_dir: ./output/re_layoutxlm/ + save_model_dir: ./output/re_layoutxlm_xfund_zh save_epoch_step: 2000 # evaluation is run every 10 iterations after the 0th iteration eval_batch_step: [ 0, 19 ] @@ -12,7 +12,7 @@ Global: use_visualdl: False seed: 2022 infer_img: ppstructure/docs/vqa/input/zh_val_21.jpg - save_res_path: ./output/re/ + save_res_path: ./output/re_layoutxlm_xfund_zh/res/ Architecture: model_type: vqa @@ -81,7 +81,7 @@ Train: loader: shuffle: True drop_last: False - batch_size_per_card: 8 + batch_size_per_card: 2 num_workers: 8 collate_fn: ListCollator diff --git a/configs/vqa/ser/layoutlm_xfund_zh.yml b/configs/kie/layoutlm_series/ser_layoutlm_xfund_zh.yml similarity index 96% rename from configs/vqa/ser/layoutlm_xfund_zh.yml rename to configs/kie/layoutlm_series/ser_layoutlm_xfund_zh.yml index 99763c1963..8c754dd8c5 100644 --- a/configs/vqa/ser/layoutlm_xfund_zh.yml +++ b/configs/kie/layoutlm_series/ser_layoutlm_xfund_zh.yml @@ -6,13 +6,13 @@ Global: save_model_dir: ./output/ser_layoutlm_xfund_zh save_epoch_step: 2000 # evaluation is run every 10 iterations after the 0th iteration - eval_batch_step: [ 0, 57 ] + eval_batch_step: [ 0, 19 ] cal_metric_during_train: False save_inference_dir: use_visualdl: False seed: 2022 infer_img: ppstructure/docs/vqa/input/zh_val_42.jpg - save_res_path: ./output/ser_layoutlm_xfund_zh/res/ + save_res_path: ./output/re_layoutlm_xfund_zh/res Architecture: model_type: vqa @@ -55,6 +55,7 @@ Train: data_dir: train_data/XFUND/zh_train/image label_file_list: - train_data/XFUND/zh_train/train.json + ratio_list: [ 1.0 ] transforms: - DecodeImage: # load image img_mode: RGB diff --git a/configs/vqa/ser/layoutlmv2_xfund_zh.yml b/configs/kie/layoutlm_series/ser_layoutlmv2_xfund_zh.yml similarity index 99% rename from configs/vqa/ser/layoutlmv2_xfund_zh.yml rename to configs/kie/layoutlm_series/ser_layoutlmv2_xfund_zh.yml index ebdc5f3169..3c0ffabe44 100644 --- a/configs/vqa/ser/layoutlmv2_xfund_zh.yml +++ b/configs/kie/layoutlm_series/ser_layoutlmv2_xfund_zh.yml @@ -27,6 +27,7 @@ Architecture: Loss: name: VQASerTokenLayoutLMLoss num_classes: *num_classes + key: "backbone_out" Optimizer: name: AdamW diff --git a/configs/vqa/ser/layoutxlm_xfund_zh.yml b/configs/kie/layoutlm_series/ser_layoutxlm_xfund_zh.yml similarity index 99% rename from configs/vqa/ser/layoutxlm_xfund_zh.yml rename to configs/kie/layoutlm_series/ser_layoutxlm_xfund_zh.yml index 68df7d9f03..18f87bdebc 100644 --- a/configs/vqa/ser/layoutxlm_xfund_zh.yml +++ b/configs/kie/layoutlm_series/ser_layoutxlm_xfund_zh.yml @@ -27,6 +27,7 @@ Architecture: Loss: name: VQASerTokenLayoutLMLoss num_classes: *num_classes + key: "backbone_out" Optimizer: name: AdamW diff --git a/configs/kie/kie_unet_sdmgr.yml b/configs/kie/sdmgr/kie_unet_sdmgr.yml similarity index 100% rename from configs/kie/kie_unet_sdmgr.yml rename to configs/kie/sdmgr/kie_unet_sdmgr.yml diff --git a/configs/vqa/re/layoutxlm_funsd.yml b/configs/kie/vi_layoutxlm/re_vi_layoutxlm_xfund_zh.yml similarity index 73% rename from configs/vqa/re/layoutxlm_funsd.yml rename to configs/kie/vi_layoutxlm/re_vi_layoutxlm_xfund_zh.yml index af28be10d6..89f7d5c3cb 100644 --- a/configs/vqa/re/layoutxlm_funsd.yml +++ b/configs/kie/vi_layoutxlm/re_vi_layoutxlm_xfund_zh.yml @@ -1,18 +1,18 @@ Global: use_gpu: True - epoch_num: &epoch_num 200 + epoch_num: &epoch_num 130 log_smooth_window: 10 print_batch_step: 10 - save_model_dir: ./output/re_layoutxlm_funsd + save_model_dir: ./output/re_vi_layoutxlm_xfund_zh save_epoch_step: 2000 # evaluation is run every 10 iterations after the 0th iteration - eval_batch_step: [ 0, 57 ] + eval_batch_step: [ 0, 19 ] cal_metric_during_train: False save_inference_dir: use_visualdl: False seed: 2022 - infer_img: train_data/FUNSD/testing_data/images/83624198.png - save_res_path: ./output/re_layoutxlm_funsd/res/ + infer_img: ppstructure/docs/vqa/input/zh_val_21.jpg + save_res_path: ./output/re/xfund_zh/with_gt Architecture: model_type: vqa @@ -21,6 +21,7 @@ Architecture: Backbone: name: LayoutXLMForRe pretrained: True + mode: vi checkpoints: Loss: @@ -50,10 +51,9 @@ Metric: Train: dataset: name: SimpleDataSet - data_dir: ./train_data/FUNSD/training_data/images/ + data_dir: train_data/XFUND/zh_train/image label_file_list: - - ./train_data/FUNSD/train_v4.json - # - ./train_data/FUNSD/train.json + - train_data/XFUND/zh_train/train.json ratio_list: [ 1.0 ] transforms: - DecodeImage: # load image @@ -62,8 +62,9 @@ Train: - VQATokenLabelEncode: # Class handling label contains_re: True algorithm: *algorithm - class_path: &class_path ./train_data/FUNSD/class_list.txt + class_path: &class_path train_data/XFUND/class_list_xfun.txt use_textline_bbox_info: &use_textline_bbox_info True + order_method: &order_method "tb-yx" - VQATokenPad: max_seq_len: &max_seq_len 512 return_attention_mask: True @@ -79,22 +80,20 @@ Train: order: 'hwc' - ToCHWImage: - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'entities', 'relations'] + keep_keys: [ 'input_ids', 'bbox','attention_mask', 'token_type_ids', 'image', 'entities', 'relations'] # dataloader will return list in this order loader: - shuffle: False + shuffle: True drop_last: False - batch_size_per_card: 8 - num_workers: 16 + batch_size_per_card: 2 + num_workers: 4 collate_fn: ListCollator Eval: dataset: name: SimpleDataSet - data_dir: ./train_data/FUNSD/testing_data/images/ - label_file_list: - - ./train_data/FUNSD/test_v4.json - # - ./train_data/FUNSD/test.json + data_dir: train_data/XFUND/zh_val/image + label_file_list: + - train_data/XFUND/zh_val/val.json transforms: - DecodeImage: # load image img_mode: RGB @@ -104,6 +103,7 @@ Eval: algorithm: *algorithm class_path: *class_path use_textline_bbox_info: *use_textline_bbox_info + order_method: *order_method - VQATokenPad: max_seq_len: *max_seq_len return_attention_mask: True @@ -119,11 +119,11 @@ Eval: order: 'hwc' - ToCHWImage: - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'entities', 'relations'] + keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'entities', 'relations'] # dataloader will return list in this order loader: shuffle: False drop_last: False batch_size_per_card: 8 num_workers: 8 collate_fn: ListCollator + diff --git a/configs/kie/vi_layoutxlm/re_vi_layoutxlm_xfund_zh_udml.yml b/configs/kie/vi_layoutxlm/re_vi_layoutxlm_xfund_zh_udml.yml new file mode 100644 index 0000000000..c1bfdb6c6c --- /dev/null +++ b/configs/kie/vi_layoutxlm/re_vi_layoutxlm_xfund_zh_udml.yml @@ -0,0 +1,175 @@ +Global: + use_gpu: True + epoch_num: &epoch_num 130 + log_smooth_window: 10 + print_batch_step: 10 + save_model_dir: ./output/re_vi_layoutxlm_xfund_zh_udml + save_epoch_step: 2000 + # evaluation is run every 10 iterations after the 0th iteration + eval_batch_step: [ 0, 19 ] + cal_metric_during_train: False + save_inference_dir: + use_visualdl: False + seed: 2022 + infer_img: ppstructure/docs/vqa/input/zh_val_21.jpg + save_res_path: ./output/re/xfund_zh/with_gt + +Architecture: + model_type: &model_type "vqa" + name: DistillationModel + algorithm: Distillation + Models: + Teacher: + pretrained: + freeze_params: false + return_all_feats: true + model_type: *model_type + algorithm: &algorithm "LayoutXLM" + Transform: + Backbone: + name: LayoutXLMForRe + pretrained: True + mode: vi + checkpoints: + Student: + pretrained: + freeze_params: false + return_all_feats: true + model_type: *model_type + algorithm: *algorithm + Transform: + Backbone: + name: LayoutXLMForRe + pretrained: True + mode: vi + checkpoints: + +Loss: + name: CombinedLoss + loss_config_list: + - DistillationLossFromOutput: + weight: 1.0 + model_name_list: ["Student", "Teacher"] + key: loss + reduction: mean + - DistillationVQADistanceLoss: + weight: 0.5 + mode: "l2" + model_name_pairs: + - ["Student", "Teacher"] + key: hidden_states_5 + name: "loss_5" + - DistillationVQADistanceLoss: + weight: 0.5 + mode: "l2" + model_name_pairs: + - ["Student", "Teacher"] + key: hidden_states_8 + name: "loss_8" + + +Optimizer: + name: AdamW + beta1: 0.9 + beta2: 0.999 + clip_norm: 10 + lr: + learning_rate: 0.00005 + warmup_epoch: 10 + regularizer: + name: L2 + factor: 0.00000 + +PostProcess: + name: DistillationRePostProcess + model_name: ["Student", "Teacher"] + key: null + + +Metric: + name: DistillationMetric + base_metric_name: VQAReTokenMetric + main_indicator: hmean + key: "Student" + +Train: + dataset: + name: SimpleDataSet + data_dir: train_data/XFUND/zh_train/image + label_file_list: + - train_data/XFUND/zh_train/train.json + ratio_list: [ 1.0 ] + transforms: + - DecodeImage: # load image + img_mode: RGB + channel_first: False + - VQATokenLabelEncode: # Class handling label + contains_re: True + algorithm: *algorithm + class_path: &class_path train_data/XFUND/class_list_xfun.txt + use_textline_bbox_info: &use_textline_bbox_info True + # [None, "tb-yx"] + order_method: &order_method "tb-yx" + - VQATokenPad: + max_seq_len: &max_seq_len 512 + return_attention_mask: True + - VQAReTokenRelation: + - VQAReTokenChunk: + max_seq_len: *max_seq_len + - Resize: + size: [224,224] + - NormalizeImage: + scale: 1 + mean: [ 123.675, 116.28, 103.53 ] + std: [ 58.395, 57.12, 57.375 ] + order: 'hwc' + - ToCHWImage: + - KeepKeys: + keep_keys: [ 'input_ids', 'bbox','attention_mask', 'token_type_ids', 'image', 'entities', 'relations'] # dataloader will return list in this order + loader: + shuffle: True + drop_last: False + batch_size_per_card: 2 + num_workers: 4 + collate_fn: ListCollator + +Eval: + dataset: + name: SimpleDataSet + data_dir: train_data/XFUND/zh_val/image + label_file_list: + - train_data/XFUND/zh_val/val.json + transforms: + - DecodeImage: # load image + img_mode: RGB + channel_first: False + - VQATokenLabelEncode: # Class handling label + contains_re: True + algorithm: *algorithm + class_path: *class_path + use_textline_bbox_info: *use_textline_bbox_info + order_method: *order_method + - VQATokenPad: + max_seq_len: *max_seq_len + return_attention_mask: True + - VQAReTokenRelation: + - VQAReTokenChunk: + max_seq_len: *max_seq_len + - Resize: + size: [224,224] + - NormalizeImage: + scale: 1 + mean: [ 123.675, 116.28, 103.53 ] + std: [ 58.395, 57.12, 57.375 ] + order: 'hwc' + - ToCHWImage: + - KeepKeys: + keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'entities', 'relations'] # dataloader will return list in this order + loader: + shuffle: False + drop_last: False + batch_size_per_card: 8 + num_workers: 8 + collate_fn: ListCollator + + diff --git a/configs/vqa/ser/layoutlm_funsd.yml b/configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh.yml similarity index 72% rename from configs/vqa/ser/layoutlm_funsd.yml rename to configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh.yml index 0ef3502bc8..d54125db64 100644 --- a/configs/vqa/ser/layoutlm_funsd.yml +++ b/configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh.yml @@ -3,30 +3,38 @@ Global: epoch_num: &epoch_num 200 log_smooth_window: 10 print_batch_step: 10 - save_model_dir: ./output/ser_layoutlm_funsd + save_model_dir: ./output/ser_vi_layoutxlm_xfund_zh save_epoch_step: 2000 # evaluation is run every 10 iterations after the 0th iteration - eval_batch_step: [ 0, 57 ] + eval_batch_step: [ 0, 19 ] cal_metric_during_train: False save_inference_dir: use_visualdl: False seed: 2022 - infer_img: train_data/FUNSD/testing_data/images/83624198.png - save_res_path: ./output/ser_layoutlm_funsd/res/ + infer_img: ppstructure/docs/vqa/input/zh_val_42.jpg + # if you want to predict using the groundtruth ocr info, + # you can use the following config + # infer_img: train_data/XFUND/zh_val/val.json + # infer_mode: False + + save_res_path: ./output/ser/xfund_zh/res Architecture: model_type: vqa - algorithm: &algorithm "LayoutLM" + algorithm: &algorithm "LayoutXLM" Transform: Backbone: - name: LayoutLMForSer + name: LayoutXLMForSer pretrained: True checkpoints: + # one of base or vi + mode: vi num_classes: &num_classes 7 Loss: name: VQASerTokenLayoutLMLoss num_classes: *num_classes + key: "backbone_out" Optimizer: name: AdamW @@ -43,7 +51,7 @@ Optimizer: PostProcess: name: VQASerTokenLayoutLMPostProcess - class_path: &class_path ./train_data/FUNSD/class_list.txt + class_path: &class_path train_data/XFUND/class_list_xfun.txt Metric: name: VQASerTokenMetric @@ -52,9 +60,10 @@ Metric: Train: dataset: name: SimpleDataSet - data_dir: ./train_data/FUNSD/training_data/images/ + data_dir: train_data/XFUND/zh_train/image label_file_list: - - ./train_data/FUNSD/train.json + - train_data/XFUND/zh_train/train.json + ratio_list: [ 1.0 ] transforms: - DecodeImage: # load image img_mode: RGB @@ -64,6 +73,8 @@ Train: algorithm: *algorithm class_path: *class_path use_textline_bbox_info: &use_textline_bbox_info True + # one of [None, "tb-yx"] + order_method: &order_method "tb-yx" - VQATokenPad: max_seq_len: &max_seq_len 512 return_attention_mask: True @@ -78,8 +89,7 @@ Train: order: 'hwc' - ToCHWImage: - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] + keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] # dataloader will return list in this order loader: shuffle: True drop_last: False @@ -89,9 +99,9 @@ Train: Eval: dataset: name: SimpleDataSet - data_dir: train_data/FUNSD/testing_data/images/ + data_dir: train_data/XFUND/zh_val/image label_file_list: - - ./train_data/FUNSD/test.json + - train_data/XFUND/zh_val/val.json transforms: - DecodeImage: # load image img_mode: RGB @@ -101,6 +111,7 @@ Eval: algorithm: *algorithm class_path: *class_path use_textline_bbox_info: *use_textline_bbox_info + order_method: *order_method - VQATokenPad: max_seq_len: *max_seq_len return_attention_mask: True @@ -115,8 +126,7 @@ Eval: order: 'hwc' - ToCHWImage: - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] + keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] # dataloader will return list in this order loader: shuffle: False drop_last: False diff --git a/configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh_udml.yml b/configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh_udml.yml new file mode 100644 index 0000000000..6f0961c8e8 --- /dev/null +++ b/configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh_udml.yml @@ -0,0 +1,183 @@ +Global: + use_gpu: True + epoch_num: &epoch_num 200 + log_smooth_window: 10 + print_batch_step: 10 + save_model_dir: ./output/ser_vi_layoutxlm_xfund_zh_udml + save_epoch_step: 2000 + # evaluation is run every 10 iterations after the 0th iteration + eval_batch_step: [ 0, 19 ] + cal_metric_during_train: False + save_inference_dir: + use_visualdl: False + seed: 2022 + infer_img: ppstructure/docs/vqa/input/zh_val_42.jpg + save_res_path: ./output/ser_layoutxlm_xfund_zh/res + + +Architecture: + model_type: &model_type "vqa" + name: DistillationModel + algorithm: Distillation + Models: + Teacher: + pretrained: + freeze_params: false + return_all_feats: true + model_type: *model_type + algorithm: &algorithm "LayoutXLM" + Transform: + Backbone: + name: LayoutXLMForSer + pretrained: True + # one of base or vi + mode: vi + checkpoints: + num_classes: &num_classes 7 + Student: + pretrained: + freeze_params: false + return_all_feats: true + model_type: *model_type + algorithm: *algorithm + Transform: + Backbone: + name: LayoutXLMForSer + pretrained: True + # one of base or vi + mode: vi + checkpoints: + num_classes: *num_classes + + +Loss: + name: CombinedLoss + loss_config_list: + - DistillationVQASerTokenLayoutLMLoss: + weight: 1.0 + model_name_list: ["Student", "Teacher"] + key: backbone_out + num_classes: *num_classes + - DistillationSERDMLLoss: + weight: 1.0 + act: "softmax" + use_log: true + model_name_pairs: + - ["Student", "Teacher"] + key: backbone_out + - DistillationVQADistanceLoss: + weight: 0.5 + mode: "l2" + model_name_pairs: + - ["Student", "Teacher"] + key: hidden_states_5 + name: "loss_5" + - DistillationVQADistanceLoss: + weight: 0.5 + mode: "l2" + model_name_pairs: + - ["Student", "Teacher"] + key: hidden_states_8 + name: "loss_8" + + + +Optimizer: + name: AdamW + beta1: 0.9 + beta2: 0.999 + lr: + name: Linear + learning_rate: 0.00005 + epochs: *epoch_num + warmup_epoch: 10 + regularizer: + name: L2 + factor: 0.00000 + +PostProcess: + name: DistillationSerPostProcess + model_name: ["Student", "Teacher"] + key: backbone_out + class_path: &class_path train_data/XFUND/class_list_xfun.txt + +Metric: + name: DistillationMetric + base_metric_name: VQASerTokenMetric + main_indicator: hmean + key: "Student" + +Train: + dataset: + name: SimpleDataSet + data_dir: train_data/XFUND/zh_train/image + label_file_list: + - train_data/XFUND/zh_train/train.json + ratio_list: [ 1.0 ] + transforms: + - DecodeImage: # load image + img_mode: RGB + channel_first: False + - VQATokenLabelEncode: # Class handling label + contains_re: False + algorithm: *algorithm + class_path: *class_path + # one of [None, "tb-yx"] + order_method: &order_method "tb-yx" + - VQATokenPad: + max_seq_len: &max_seq_len 512 + return_attention_mask: True + - VQASerTokenChunk: + max_seq_len: *max_seq_len + - Resize: + size: [224,224] + - NormalizeImage: + scale: 1 + mean: [ 123.675, 116.28, 103.53 ] + std: [ 58.395, 57.12, 57.375 ] + order: 'hwc' + - ToCHWImage: + - KeepKeys: + keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] # dataloader will return list in this order + loader: + shuffle: True + drop_last: False + batch_size_per_card: 4 + num_workers: 4 + +Eval: + dataset: + name: SimpleDataSet + data_dir: train_data/XFUND/zh_val/image + label_file_list: + - train_data/XFUND/zh_val/val.json + transforms: + - DecodeImage: # load image + img_mode: RGB + channel_first: False + - VQATokenLabelEncode: # Class handling label + contains_re: False + algorithm: *algorithm + class_path: *class_path + order_method: *order_method + - VQATokenPad: + max_seq_len: *max_seq_len + return_attention_mask: True + - VQASerTokenChunk: + max_seq_len: *max_seq_len + - Resize: + size: [224,224] + - NormalizeImage: + scale: 1 + mean: [ 123.675, 116.28, 103.53 ] + std: [ 58.395, 57.12, 57.375 ] + order: 'hwc' + - ToCHWImage: + - KeepKeys: + keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] # dataloader will return list in this order + loader: + shuffle: False + drop_last: False + batch_size_per_card: 8 + num_workers: 4 + diff --git a/configs/vqa/re/layoutlmv2_funsd.yml b/configs/vqa/re/layoutlmv2_funsd.yml deleted file mode 100644 index 1c3d8f7854..0000000000 --- a/configs/vqa/re/layoutlmv2_funsd.yml +++ /dev/null @@ -1,125 +0,0 @@ -Global: - use_gpu: True - epoch_num: &epoch_num 200 - log_smooth_window: 10 - print_batch_step: 10 - save_model_dir: ./output/re_layoutlmv2_funsd - save_epoch_step: 2000 - # evaluation is run every 10 iterations after the 0th iteration - eval_batch_step: [ 0, 57 ] - cal_metric_during_train: False - save_inference_dir: - use_visualdl: False - seed: 2022 - infer_img: train_data/FUNSD/testing_data/images/83624198.png - save_res_path: ./output/re_layoutlmv2_funsd/res/ - -Architecture: - model_type: vqa - algorithm: &algorithm "LayoutLMv2" - Transform: - Backbone: - name: LayoutLMv2ForRe - pretrained: True - checkpoints: - -Loss: - name: LossFromOutput - key: loss - reduction: mean - -Optimizer: - name: AdamW - beta1: 0.9 - beta2: 0.999 - clip_norm: 10 - lr: - learning_rate: 0.00005 - warmup_epoch: 10 - regularizer: - name: L2 - factor: 0.00000 - -PostProcess: - name: VQAReTokenLayoutLMPostProcess - -Metric: - name: VQAReTokenMetric - main_indicator: hmean - -Train: - dataset: - name: SimpleDataSet - data_dir: ./train_data/FUNSD/training_data/images/ - label_file_list: - - ./train_data/FUNSD/train.json - ratio_list: [ 1.0 ] - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: True - algorithm: *algorithm - class_path: &class_path train_data/FUNSD/class_list.txt - - VQATokenPad: - max_seq_len: &max_seq_len 512 - return_attention_mask: True - - VQAReTokenRelation: - - VQAReTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1./255. - mean: [0.485, 0.456, 0.406] - std: [0.229, 0.224, 0.225] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'entities', 'relations'] - loader: - shuffle: True - drop_last: False - batch_size_per_card: 8 - num_workers: 8 - collate_fn: ListCollator - -Eval: - dataset: - name: SimpleDataSet - data_dir: ./train_data/FUNSD/testing_data/images/ - label_file_list: - - ./train_data/FUNSD/test.json - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: True - algorithm: *algorithm - class_path: *class_path - - VQATokenPad: - max_seq_len: *max_seq_len - return_attention_mask: True - - VQAReTokenRelation: - - VQAReTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1./255. - mean: [0.485, 0.456, 0.406] - std: [0.229, 0.224, 0.225] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'entities', 'relations'] - loader: - shuffle: False - drop_last: False - batch_size_per_card: 8 - num_workers: 8 - collate_fn: ListCollator diff --git a/configs/vqa/ser/layoutlm_sroie.yml b/configs/vqa/ser/layoutlm_sroie.yml deleted file mode 100644 index 6abb1151e5..0000000000 --- a/configs/vqa/ser/layoutlm_sroie.yml +++ /dev/null @@ -1,124 +0,0 @@ -Global: - use_gpu: True - epoch_num: &epoch_num 200 - log_smooth_window: 10 - print_batch_step: 10 - save_model_dir: ./output/ser_layoutlm_sroie - save_epoch_step: 2000 - # evaluation is run every 10 iterations after the 0th iteration - eval_batch_step: [ 0, 200 ] - cal_metric_during_train: False - save_inference_dir: - use_visualdl: False - seed: 2022 - infer_img: train_data/SROIE/test/X00016469670.jpg - save_res_path: ./output/ser_layoutlm_sroie/res/ - -Architecture: - model_type: vqa - algorithm: &algorithm "LayoutLM" - Transform: - Backbone: - name: LayoutLMForSer - pretrained: True - checkpoints: - num_classes: &num_classes 9 - -Loss: - name: VQASerTokenLayoutLMLoss - num_classes: *num_classes - -Optimizer: - name: AdamW - beta1: 0.9 - beta2: 0.999 - lr: - name: Linear - learning_rate: 0.00005 - epochs: *epoch_num - warmup_epoch: 2 - regularizer: - name: L2 - factor: 0.00000 - -PostProcess: - name: VQASerTokenLayoutLMPostProcess - class_path: &class_path ./train_data/SROIE/class_list.txt - -Metric: - name: VQASerTokenMetric - main_indicator: hmean - -Train: - dataset: - name: SimpleDataSet - data_dir: ./train_data/SROIE/train - label_file_list: - - ./train_data/SROIE/train.txt - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: False - algorithm: *algorithm - class_path: *class_path - use_textline_bbox_info: &use_textline_bbox_info True - - VQATokenPad: - max_seq_len: &max_seq_len 512 - return_attention_mask: True - - VQASerTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1 - mean: [ 123.675, 116.28, 103.53 ] - std: [ 58.395, 57.12, 57.375 ] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] - loader: - shuffle: True - drop_last: False - batch_size_per_card: 8 - num_workers: 4 - -Eval: - dataset: - name: SimpleDataSet - data_dir: ./train_data/SROIE/test - label_file_list: - - ./train_data/SROIE/test.txt - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: False - algorithm: *algorithm - class_path: *class_path - use_textline_bbox_info: *use_textline_bbox_info - - VQATokenPad: - max_seq_len: *max_seq_len - return_attention_mask: True - - VQASerTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1 - mean: [ 123.675, 116.28, 103.53 ] - std: [ 58.395, 57.12, 57.375 ] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] - loader: - shuffle: False - drop_last: False - batch_size_per_card: 8 - num_workers: 4 diff --git a/configs/vqa/ser/layoutlmv2_funsd.yml b/configs/vqa/ser/layoutlmv2_funsd.yml deleted file mode 100644 index 438edc1aa7..0000000000 --- a/configs/vqa/ser/layoutlmv2_funsd.yml +++ /dev/null @@ -1,123 +0,0 @@ -Global: - use_gpu: True - epoch_num: &epoch_num 200 - log_smooth_window: 10 - print_batch_step: 10 - save_model_dir: ./output/ser_layoutlmv2_funsd - save_epoch_step: 2000 - # evaluation is run every 10 iterations after the 0th iteration - eval_batch_step: [ 0, 100 ] - cal_metric_during_train: False - save_inference_dir: - use_visualdl: False - seed: 2022 - infer_img: train_data/FUNSD/testing_data/images/83624198.png - save_res_path: ./output/ser_layoutlmv2_funsd/res/ - -Architecture: - model_type: vqa - algorithm: &algorithm "LayoutLMv2" - Transform: - Backbone: - name: LayoutLMv2ForSer - pretrained: True - checkpoints: - num_classes: &num_classes 7 - -Loss: - name: VQASerTokenLayoutLMLoss - num_classes: *num_classes - -Optimizer: - name: AdamW - beta1: 0.9 - beta2: 0.999 - lr: - name: Linear - learning_rate: 0.00005 - epochs: *epoch_num - warmup_epoch: 2 - regularizer: - - name: L2 - factor: 0.00000 - -PostProcess: - name: VQASerTokenLayoutLMPostProcess - class_path: &class_path train_data/FUNSD/class_list.txt - -Metric: - name: VQASerTokenMetric - main_indicator: hmean - -Train: - dataset: - name: SimpleDataSet - data_dir: ./train_data/FUNSD/training_data/images/ - label_file_list: - - ./train_data/FUNSD/train.json - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: False - algorithm: *algorithm - class_path: *class_path - - VQATokenPad: - max_seq_len: &max_seq_len 512 - return_attention_mask: True - - VQASerTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1 - mean: [ 123.675, 116.28, 103.53 ] - std: [ 58.395, 57.12, 57.375 ] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] - loader: - shuffle: True - drop_last: False - batch_size_per_card: 8 - num_workers: 4 - -Eval: - dataset: - name: SimpleDataSet - data_dir: ./train_data/FUNSD/testing_data/images/ - label_file_list: - - ./train_data/FUNSD/test.json - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: False - algorithm: *algorithm - class_path: *class_path - - VQATokenPad: - max_seq_len: *max_seq_len - return_attention_mask: True - - VQASerTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1 - mean: [ 123.675, 116.28, 103.53 ] - std: [ 58.395, 57.12, 57.375 ] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] - loader: - shuffle: False - drop_last: False - batch_size_per_card: 8 - num_workers: 4 diff --git a/configs/vqa/ser/layoutlmv2_sroie.yml b/configs/vqa/ser/layoutlmv2_sroie.yml deleted file mode 100644 index 549beb8ec5..0000000000 --- a/configs/vqa/ser/layoutlmv2_sroie.yml +++ /dev/null @@ -1,123 +0,0 @@ -Global: - use_gpu: True - epoch_num: &epoch_num 200 - log_smooth_window: 10 - print_batch_step: 10 - save_model_dir: ./output/ser_layoutlmv2_sroie - save_epoch_step: 2000 - # evaluation is run every 10 iterations after the 0th iteration - eval_batch_step: [ 0, 200 ] - cal_metric_during_train: False - save_inference_dir: - use_visualdl: False - seed: 2022 - infer_img: train_data/SROIE/test/X00016469670.jpg - save_res_path: ./output/ser_layoutlmv2_sroie/res/ - -Architecture: - model_type: vqa - algorithm: &algorithm "LayoutLMv2" - Transform: - Backbone: - name: LayoutLMv2ForSer - pretrained: True - checkpoints: - num_classes: &num_classes 9 - -Loss: - name: VQASerTokenLayoutLMLoss - num_classes: *num_classes - -Optimizer: - name: AdamW - beta1: 0.9 - beta2: 0.999 - lr: - name: Linear - learning_rate: 0.00005 - epochs: *epoch_num - warmup_epoch: 2 - regularizer: - - name: L2 - factor: 0.00000 - -PostProcess: - name: VQASerTokenLayoutLMPostProcess - class_path: &class_path ./train_data/SROIE/class_list.txt - -Metric: - name: VQASerTokenMetric - main_indicator: hmean - -Train: - dataset: - name: SimpleDataSet - data_dir: ./train_data/SROIE/train - label_file_list: - - ./train_data/SROIE/train.txt - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: False - algorithm: *algorithm - class_path: *class_path - - VQATokenPad: - max_seq_len: &max_seq_len 512 - return_attention_mask: True - - VQASerTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1 - mean: [ 123.675, 116.28, 103.53 ] - std: [ 58.395, 57.12, 57.375 ] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] - loader: - shuffle: True - drop_last: False - batch_size_per_card: 8 - num_workers: 4 - -Eval: - dataset: - name: SimpleDataSet - data_dir: ./train_data/SROIE/test - label_file_list: - - ./train_data/SROIE/test.txt - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: False - algorithm: *algorithm - class_path: *class_path - - VQATokenPad: - max_seq_len: *max_seq_len - return_attention_mask: True - - VQASerTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1 - mean: [ 123.675, 116.28, 103.53 ] - std: [ 58.395, 57.12, 57.375 ] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] - loader: - shuffle: False - drop_last: False - batch_size_per_card: 8 - num_workers: 4 diff --git a/configs/vqa/ser/layoutxlm_funsd.yml b/configs/vqa/ser/layoutxlm_funsd.yml deleted file mode 100644 index be1e9d4f1e..0000000000 --- a/configs/vqa/ser/layoutxlm_funsd.yml +++ /dev/null @@ -1,123 +0,0 @@ -Global: - use_gpu: True - epoch_num: &epoch_num 200 - log_smooth_window: 10 - print_batch_step: 10 - save_model_dir: ./output/ser_layoutxlm_funsd - save_epoch_step: 2000 - # evaluation is run every 10 iterations after the 0th iteration - eval_batch_step: [ 0, 57 ] - cal_metric_during_train: False - save_inference_dir: - use_visualdl: False - seed: 2022 - infer_img: train_data/FUNSD/testing_data/images/83624198.png - save_res_path: output/ser_layoutxlm_funsd/res/ - -Architecture: - model_type: vqa - algorithm: &algorithm "LayoutXLM" - Transform: - Backbone: - name: LayoutXLMForSer - pretrained: True - checkpoints: - num_classes: &num_classes 7 - -Loss: - name: VQASerTokenLayoutLMLoss - num_classes: *num_classes - -Optimizer: - name: AdamW - beta1: 0.9 - beta2: 0.999 - lr: - name: Linear - learning_rate: 0.00005 - epochs: *epoch_num - warmup_epoch: 2 - regularizer: - name: L2 - factor: 0.00000 - -PostProcess: - name: VQASerTokenLayoutLMPostProcess - class_path: &class_path ./train_data/FUNSD/class_list.txt - -Metric: - name: VQASerTokenMetric - main_indicator: hmean - -Train: - dataset: - name: SimpleDataSet - data_dir: ./train_data/FUNSD/training_data/images/ - label_file_list: - - ./train_data/FUNSD/train.json - ratio_list: [ 1.0 ] - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: False - algorithm: *algorithm - class_path: *class_path - - VQATokenPad: - max_seq_len: &max_seq_len 512 - return_attention_mask: True - - VQASerTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1 - mean: [ 123.675, 116.28, 103.53 ] - std: [ 58.395, 57.12, 57.375 ] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] - loader: - shuffle: True - drop_last: False - batch_size_per_card: 8 - num_workers: 4 - -Eval: - dataset: - name: SimpleDataSet - data_dir: train_data/FUNSD/testing_data/images/ - label_file_list: - - ./train_data/FUNSD/test.json - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: False - algorithm: *algorithm - class_path: *class_path - - VQATokenPad: - max_seq_len: *max_seq_len - return_attention_mask: True - - VQASerTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1 - mean: [ 123.675, 116.28, 103.53 ] - std: [ 58.395, 57.12, 57.375 ] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] - loader: - shuffle: False - drop_last: False - batch_size_per_card: 8 - num_workers: 4 diff --git a/configs/vqa/ser/layoutxlm_sroie.yml b/configs/vqa/ser/layoutxlm_sroie.yml deleted file mode 100644 index dd63d888d0..0000000000 --- a/configs/vqa/ser/layoutxlm_sroie.yml +++ /dev/null @@ -1,123 +0,0 @@ -Global: - use_gpu: True - epoch_num: &epoch_num 200 - log_smooth_window: 10 - print_batch_step: 10 - save_model_dir: ./output/ser_layoutxlm_sroie - save_epoch_step: 2000 - # evaluation is run every 10 iterations after the 0th iteration - eval_batch_step: [ 0, 200 ] - cal_metric_during_train: False - save_inference_dir: - use_visualdl: False - seed: 2022 - infer_img: train_data/SROIE/test/X00016469670.jpg - save_res_path: res_img_aug_with_gt - -Architecture: - model_type: vqa - algorithm: &algorithm "LayoutXLM" - Transform: - Backbone: - name: LayoutXLMForSer - pretrained: True - checkpoints: - num_classes: &num_classes 9 - -Loss: - name: VQASerTokenLayoutLMLoss - num_classes: *num_classes - -Optimizer: - name: AdamW - beta1: 0.9 - beta2: 0.999 - lr: - name: Linear - learning_rate: 0.00005 - epochs: *epoch_num - warmup_epoch: 2 - regularizer: - name: L2 - factor: 0.00000 - -PostProcess: - name: VQASerTokenLayoutLMPostProcess - class_path: &class_path ./train_data/SROIE/class_list.txt - -Metric: - name: VQASerTokenMetric - main_indicator: hmean - -Train: - dataset: - name: SimpleDataSet - data_dir: ./train_data/SROIE/train - label_file_list: - - ./train_data/SROIE/train.txt - ratio_list: [ 1.0 ] - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: False - algorithm: *algorithm - class_path: *class_path - - VQATokenPad: - max_seq_len: &max_seq_len 512 - return_attention_mask: True - - VQASerTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1 - mean: [ 123.675, 116.28, 103.53 ] - std: [ 58.395, 57.12, 57.375 ] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] - loader: - shuffle: True - drop_last: False - batch_size_per_card: 8 - num_workers: 4 - -Eval: - dataset: - name: SimpleDataSet - data_dir: train_data/SROIE/test - label_file_list: - - ./train_data/SROIE/test.txt - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: False - algorithm: *algorithm - class_path: *class_path - - VQATokenPad: - max_seq_len: *max_seq_len - return_attention_mask: True - - VQASerTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1 - mean: [ 123.675, 116.28, 103.53 ] - std: [ 58.395, 57.12, 57.375 ] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] - loader: - shuffle: False - drop_last: False - batch_size_per_card: 8 - num_workers: 4 diff --git a/configs/vqa/ser/layoutxlm_wildreceipt.yml b/configs/vqa/ser/layoutxlm_wildreceipt.yml deleted file mode 100644 index 92c0394296..0000000000 --- a/configs/vqa/ser/layoutxlm_wildreceipt.yml +++ /dev/null @@ -1,123 +0,0 @@ -Global: - use_gpu: True - epoch_num: &epoch_num 100 - log_smooth_window: 10 - print_batch_step: 10 - save_model_dir: ./output/ser_layoutxlm_wildreceipt - save_epoch_step: 2000 - # evaluation is run every 10 iterations after the 0th iteration - eval_batch_step: [ 0, 200 ] - cal_metric_during_train: False - save_inference_dir: - use_visualdl: False - seed: 2022 - infer_img: train_data//wildreceipt/image_files/Image_12/10/845be0dd6f5b04866a2042abd28d558032ef2576.jpeg - save_res_path: ./output/ser_layoutxlm_wildreceipt/res - -Architecture: - model_type: vqa - algorithm: &algorithm "LayoutXLM" - Transform: - Backbone: - name: LayoutXLMForSer - pretrained: True - checkpoints: - num_classes: &num_classes 51 - -Loss: - name: VQASerTokenLayoutLMLoss - num_classes: *num_classes - -Optimizer: - name: AdamW - beta1: 0.9 - beta2: 0.999 - lr: - name: Linear - learning_rate: 0.00005 - epochs: *epoch_num - warmup_epoch: 2 - regularizer: - name: L2 - factor: 0.00000 - -PostProcess: - name: VQASerTokenLayoutLMPostProcess - class_path: &class_path ./train_data/wildreceipt/class_list.txt - -Metric: - name: VQASerTokenMetric - main_indicator: hmean - -Train: - dataset: - name: SimpleDataSet - data_dir: ./train_data/wildreceipt/ - label_file_list: - - ./train_data/wildreceipt/wildreceipt_train.txt - ratio_list: [ 1.0 ] - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: False - algorithm: *algorithm - class_path: *class_path - - VQATokenPad: - max_seq_len: &max_seq_len 512 - return_attention_mask: True - - VQASerTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1 - mean: [ 123.675, 116.28, 103.53 ] - std: [ 58.395, 57.12, 57.375 ] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] - loader: - shuffle: True - drop_last: False - batch_size_per_card: 8 - num_workers: 4 - -Eval: - dataset: - name: SimpleDataSet - data_dir: train_data/wildreceipt - label_file_list: - - ./train_data/wildreceipt/wildreceipt_test.txt - transforms: - - DecodeImage: # load image - img_mode: RGB - channel_first: False - - VQATokenLabelEncode: # Class handling label - contains_re: False - algorithm: *algorithm - class_path: *class_path - - VQATokenPad: - max_seq_len: *max_seq_len - return_attention_mask: True - - VQASerTokenChunk: - max_seq_len: *max_seq_len - - Resize: - size: [224,224] - - NormalizeImage: - scale: 1 - mean: [ 123.675, 116.28, 103.53 ] - std: [ 58.395, 57.12, 57.375 ] - order: 'hwc' - - ToCHWImage: - - KeepKeys: - # dataloader will return list in this order - keep_keys: [ 'input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] - loader: - shuffle: False - drop_last: False - batch_size_per_card: 8 - num_workers: 4 diff --git a/ppocr/data/imaug/label_ops.py b/ppocr/data/imaug/label_ops.py index d84f966201..180fe97b47 100644 --- a/ppocr/data/imaug/label_ops.py +++ b/ppocr/data/imaug/label_ops.py @@ -24,6 +24,7 @@ from shapely.geometry import LineString, Point, Polygon import json import copy from ppocr.utils.logging import get_logger +from ppocr.data.imaug.vqa.augment import order_by_tbyx class ClsLabelEncode(object): @@ -870,6 +871,7 @@ class VQATokenLabelEncode(object): add_special_ids=False, algorithm='LayoutXLM', use_textline_bbox_info=True, + order_method=None, infer_mode=False, ocr_engine=None, **kwargs): @@ -899,6 +901,8 @@ class VQATokenLabelEncode(object): self.infer_mode = infer_mode self.ocr_engine = ocr_engine self.use_textline_bbox_info = use_textline_bbox_info + self.order_method = order_method + assert self.order_method in [None, "tb-yx"] def split_bbox(self, bbox, text, tokenizer): words = text.split() @@ -938,6 +942,14 @@ class VQATokenLabelEncode(object): # load bbox and label info ocr_info = self._load_ocr_info(data) + for idx in range(len(ocr_info)): + if "bbox" not in ocr_info[idx]: + ocr_info[idx]["bbox"] = self.trans_poly_to_bbox(ocr_info[idx][ + "points"]) + + if self.order_method == "tb-yx": + ocr_info = order_by_tbyx(ocr_info) + # for re train_re = self.contains_re and not self.infer_mode if train_re: @@ -1052,10 +1064,10 @@ class VQATokenLabelEncode(object): return data def trans_poly_to_bbox(self, poly): - x1 = np.min([p[0] for p in poly]) - x2 = np.max([p[0] for p in poly]) - y1 = np.min([p[1] for p in poly]) - y2 = np.max([p[1] for p in poly]) + x1 = int(np.min([p[0] for p in poly])) + x2 = int(np.max([p[0] for p in poly])) + y1 = int(np.min([p[1] for p in poly])) + y2 = int(np.max([p[1] for p in poly])) return [x1, y1, x2, y2] def _load_ocr_info(self, data): diff --git a/ppocr/data/imaug/vqa/__init__.py b/ppocr/data/imaug/vqa/__init__.py index bde1751155..34189bcefb 100644 --- a/ppocr/data/imaug/vqa/__init__.py +++ b/ppocr/data/imaug/vqa/__init__.py @@ -13,12 +13,10 @@ # limitations under the License. from .token import VQATokenPad, VQASerTokenChunk, VQAReTokenChunk, VQAReTokenRelation -from .augment import DistortBBox __all__ = [ 'VQATokenPad', 'VQASerTokenChunk', 'VQAReTokenChunk', 'VQAReTokenRelation', - 'DistortBBox', ] diff --git a/ppocr/data/imaug/vqa/augment.py b/ppocr/data/imaug/vqa/augment.py index fcdc9685e9..b95fcdf0f0 100644 --- a/ppocr/data/imaug/vqa/augment.py +++ b/ppocr/data/imaug/vqa/augment.py @@ -16,22 +16,18 @@ import os import sys import numpy as np import random +from copy import deepcopy -class DistortBBox: - def __init__(self, prob=0.5, max_scale=1, **kwargs): - """Random distort bbox - """ - self.prob = prob - self.max_scale = max_scale - - def __call__(self, data): - if random.random() > self.prob: - return data - bbox = np.array(data['bbox']) - rnd_scale = (np.random.rand(*bbox.shape) - 0.5) * 2 * self.max_scale - bbox = np.round(bbox + rnd_scale).astype(bbox.dtype) - data['bbox'] = np.clip(data['bbox'], 0, 1000) - data['bbox'] = bbox.tolist() - sys.stdout.flush() - return data +def order_by_tbyx(ocr_info): + res = sorted(ocr_info, key=lambda r: (r["bbox"][1], r["bbox"][0])) + for i in range(len(res) - 1): + for j in range(i, 0, -1): + if abs(res[j + 1]["bbox"][1] - res[j]["bbox"][1]) < 20 and \ + (res[j + 1]["bbox"][0] < res[j]["bbox"][0]): + tmp = deepcopy(res[j]) + res[j] = deepcopy(res[j + 1]) + res[j + 1] = deepcopy(tmp) + else: + break + return res diff --git a/ppocr/losses/basic_loss.py b/ppocr/losses/basic_loss.py index 74490791c2..da9faa08bc 100644 --- a/ppocr/losses/basic_loss.py +++ b/ppocr/losses/basic_loss.py @@ -63,18 +63,21 @@ class KLJSLoss(object): def __call__(self, p1, p2, reduction="mean"): if self.mode.lower() == 'kl': - loss = paddle.multiply(p2, paddle.log((p2 + 1e-5) / (p1 + 1e-5) + 1e-5)) + loss = paddle.multiply(p2, + paddle.log((p2 + 1e-5) / (p1 + 1e-5) + 1e-5)) loss += paddle.multiply( - p1, paddle.log((p1 + 1e-5) / (p2 + 1e-5) + 1e-5)) + p1, paddle.log((p1 + 1e-5) / (p2 + 1e-5) + 1e-5)) loss *= 0.5 elif self.mode.lower() == "js": - loss = paddle.multiply(p2, paddle.log((2*p2 + 1e-5) / (p1 + p2 + 1e-5) + 1e-5)) + loss = paddle.multiply( + p2, paddle.log((2 * p2 + 1e-5) / (p1 + p2 + 1e-5) + 1e-5)) loss += paddle.multiply( - p1, paddle.log((2*p1 + 1e-5) / (p1 + p2 + 1e-5) + 1e-5)) + p1, paddle.log((2 * p1 + 1e-5) / (p1 + p2 + 1e-5) + 1e-5)) loss *= 0.5 else: - raise ValueError("The mode.lower() if KLJSLoss should be one of ['kl', 'js']") - + raise ValueError( + "The mode.lower() if KLJSLoss should be one of ['kl', 'js']") + if reduction == "mean": loss = paddle.mean(loss, axis=[1, 2]) elif reduction == "none" or reduction is None: @@ -154,7 +157,9 @@ class LossFromOutput(nn.Layer): self.reduction = reduction def forward(self, predicts, batch): - loss = predicts[self.key] + loss = predicts + if self.key is not None and isinstance(predicts, dict): + loss = loss[self.key] if self.reduction == 'mean': loss = paddle.mean(loss) elif self.reduction == 'sum': diff --git a/ppocr/losses/combined_loss.py b/ppocr/losses/combined_loss.py index f4cdee8f90..8d697d544b 100644 --- a/ppocr/losses/combined_loss.py +++ b/ppocr/losses/combined_loss.py @@ -24,6 +24,9 @@ from .distillation_loss import DistillationCTCLoss from .distillation_loss import DistillationSARLoss from .distillation_loss import DistillationDMLLoss from .distillation_loss import DistillationDistanceLoss, DistillationDBLoss, DistillationDilaDBLoss +from .distillation_loss import DistillationVQASerTokenLayoutLMLoss, DistillationSERDMLLoss +from .distillation_loss import DistillationLossFromOutput +from .distillation_loss import DistillationVQADistanceLoss class CombinedLoss(nn.Layer): diff --git a/ppocr/losses/distillation_loss.py b/ppocr/losses/distillation_loss.py index 565b066d13..87fed6235d 100644 --- a/ppocr/losses/distillation_loss.py +++ b/ppocr/losses/distillation_loss.py @@ -21,8 +21,10 @@ from .rec_ctc_loss import CTCLoss from .rec_sar_loss import SARLoss from .basic_loss import DMLLoss from .basic_loss import DistanceLoss +from .basic_loss import LossFromOutput from .det_db_loss import DBLoss from .det_basic_loss import BalanceLoss, MaskL1Loss, DiceLoss +from .vqa_token_layoutlm_loss import VQASerTokenLayoutLMLoss def _sum_loss(loss_dict): @@ -322,3 +324,133 @@ class DistillationDistanceLoss(DistanceLoss): loss_dict["{}_{}_{}_{}".format(self.name, pair[0], pair[1], idx)] = loss return loss_dict + + +class DistillationVQASerTokenLayoutLMLoss(VQASerTokenLayoutLMLoss): + def __init__(self, + num_classes, + model_name_list=[], + key=None, + name="loss_ser"): + super().__init__(num_classes=num_classes) + self.model_name_list = model_name_list + self.key = key + self.name = name + + def forward(self, predicts, batch): + loss_dict = dict() + for idx, model_name in enumerate(self.model_name_list): + out = predicts[model_name] + if self.key is not None: + out = out[self.key] + loss = super().forward(out, batch) + loss_dict["{}_{}".format(self.name, model_name)] = loss["loss"] + return loss_dict + + +class DistillationLossFromOutput(LossFromOutput): + def __init__(self, + reduction="none", + model_name_list=[], + dist_key=None, + key="loss", + name="loss_re"): + super().__init__(key=key, reduction=reduction) + self.model_name_list = model_name_list + self.name = name + self.dist_key = dist_key + + def forward(self, predicts, batch): + loss_dict = dict() + for idx, model_name in enumerate(self.model_name_list): + out = predicts[model_name] + if self.dist_key is not None: + out = out[self.dist_key] + loss = super().forward(out, batch) + loss_dict["{}_{}".format(self.name, model_name)] = loss["loss"] + return loss_dict + + +class DistillationSERDMLLoss(DMLLoss): + """ + """ + + def __init__(self, + act="softmax", + use_log=True, + num_classes=7, + model_name_pairs=[], + key=None, + name="loss_dml_ser"): + super().__init__(act=act, use_log=use_log) + assert isinstance(model_name_pairs, list) + self.key = key + self.name = name + self.num_classes = num_classes + self.model_name_pairs = model_name_pairs + + def forward(self, predicts, batch): + loss_dict = dict() + for idx, pair in enumerate(self.model_name_pairs): + out1 = predicts[pair[0]] + out2 = predicts[pair[1]] + if self.key is not None: + out1 = out1[self.key] + out2 = out2[self.key] + out1 = out1.reshape([-1, out1.shape[-1]]) + out2 = out2.reshape([-1, out2.shape[-1]]) + + attention_mask = batch[2] + if attention_mask is not None: + active_output = attention_mask.reshape([-1, ]) == 1 + out1 = out1[active_output] + out2 = out2[active_output] + + loss_dict["{}_{}".format(self.name, idx)] = super().forward(out1, + out2) + + return loss_dict + + +class DistillationVQADistanceLoss(DistanceLoss): + def __init__(self, + mode="l2", + model_name_pairs=[], + key=None, + name="loss_distance", + **kargs): + super().__init__(mode=mode, **kargs) + assert isinstance(model_name_pairs, list) + self.key = key + self.model_name_pairs = model_name_pairs + self.name = name + "_l2" + + def forward(self, predicts, batch): + loss_dict = dict() + for idx, pair in enumerate(self.model_name_pairs): + out1 = predicts[pair[0]] + out2 = predicts[pair[1]] + attention_mask = batch[2] + if self.key is not None: + out1 = out1[self.key] + out2 = out2[self.key] + if attention_mask is not None: + max_len = attention_mask.shape[-1] + out1 = out1[:, :max_len] + out2 = out2[:, :max_len] + out1 = out1.reshape([-1, out1.shape[-1]]) + out2 = out2.reshape([-1, out2.shape[-1]]) + if attention_mask is not None: + active_output = attention_mask.reshape([-1, ]) == 1 + out1 = out1[active_output] + out2 = out2[active_output] + + loss = super().forward(out1, out2) + if isinstance(loss, dict): + for key in loss: + loss_dict["{}_{}nohu_{}".format(self.name, key, + idx)] = loss[key] + else: + loss_dict["{}_{}_{}_{}".format(self.name, pair[0], pair[1], + idx)] = loss + return loss_dict diff --git a/ppocr/losses/vqa_token_layoutlm_loss.py b/ppocr/losses/vqa_token_layoutlm_loss.py index f9cd463473..5d564c0e26 100755 --- a/ppocr/losses/vqa_token_layoutlm_loss.py +++ b/ppocr/losses/vqa_token_layoutlm_loss.py @@ -17,26 +17,30 @@ from __future__ import division from __future__ import print_function from paddle import nn +from ppocr.losses.basic_loss import DMLLoss class VQASerTokenLayoutLMLoss(nn.Layer): - def __init__(self, num_classes): + def __init__(self, num_classes, key=None): super().__init__() self.loss_class = nn.CrossEntropyLoss() self.num_classes = num_classes self.ignore_index = self.loss_class.ignore_index + self.key = key def forward(self, predicts, batch): + if isinstance(predicts, dict) and self.key is not None: + predicts = predicts[self.key] labels = batch[5] attention_mask = batch[2] if attention_mask is not None: active_loss = attention_mask.reshape([-1, ]) == 1 - active_outputs = predicts.reshape( + active_output = predicts.reshape( [-1, self.num_classes])[active_loss] - active_labels = labels.reshape([-1, ])[active_loss] - loss = self.loss_class(active_outputs, active_labels) + active_label = labels.reshape([-1, ])[active_loss] + loss = self.loss_class(active_output, active_label) else: loss = self.loss_class( predicts.reshape([-1, self.num_classes]), labels.reshape([-1, ])) - return {'loss': loss} + return {'loss': loss} \ No newline at end of file diff --git a/ppocr/metrics/distillation_metric.py b/ppocr/metrics/distillation_metric.py index c440cebdd0..e2cbc4dc07 100644 --- a/ppocr/metrics/distillation_metric.py +++ b/ppocr/metrics/distillation_metric.py @@ -19,6 +19,8 @@ from .rec_metric import RecMetric from .det_metric import DetMetric from .e2e_metric import E2EMetric from .cls_metric import ClsMetric +from .vqa_token_ser_metric import VQASerTokenMetric +from .vqa_token_re_metric import VQAReTokenMetric class DistillationMetric(object): diff --git a/ppocr/modeling/architectures/base_model.py b/ppocr/modeling/architectures/base_model.py index c6b50d4886..ed2a909cb5 100644 --- a/ppocr/modeling/architectures/base_model.py +++ b/ppocr/modeling/architectures/base_model.py @@ -73,28 +73,40 @@ class BaseModel(nn.Layer): self.return_all_feats = config.get("return_all_feats", False) def forward(self, x, data=None): + y = dict() if self.use_transform: x = self.transform(x) x = self.backbone(x) - y["backbone_out"] = x - if self.use_neck: - x = self.neck(x) - y["neck_out"] = x - if self.use_head: - x = self.head(x, targets=data) - # for multi head, save ctc neck out for udml - if isinstance(x, dict) and 'ctc_neck' in x.keys(): - y["neck_out"] = x["ctc_neck"] - y["head_out"] = x - elif isinstance(x, dict): + if isinstance(x, dict): y.update(x) else: - y["head_out"] = x + y["backbone_out"] = x + final_name = "backbone_out" + if self.use_neck: + x = self.neck(x) + if isinstance(x, dict): + y.update(x) + else: + y["neck_out"] = x + final_name = "neck_out" + if self.use_head: + x = self.head(x, targets=data) + # for multi head, save ctc neck out for udml + if isinstance(x, dict) and 'ctc_neck' in x.keys(): + y["neck_out"] = x["ctc_neck"] + y["head_out"] = x + elif isinstance(x, dict): + y.update(x) + else: + y["head_out"] = x + final_name = "head_out" if self.return_all_feats: if self.training: return y + elif isinstance(x, dict): + return x else: - return {"head_out": y["head_out"]} + return {final_name: x} else: return x diff --git a/ppocr/modeling/backbones/vqa_layoutlm.py b/ppocr/modeling/backbones/vqa_layoutlm.py index 34dd9d10ea..d4ced35088 100644 --- a/ppocr/modeling/backbones/vqa_layoutlm.py +++ b/ppocr/modeling/backbones/vqa_layoutlm.py @@ -22,13 +22,22 @@ from paddle import nn from paddlenlp.transformers import LayoutXLMModel, LayoutXLMForTokenClassification, LayoutXLMForRelationExtraction from paddlenlp.transformers import LayoutLMModel, LayoutLMForTokenClassification from paddlenlp.transformers import LayoutLMv2Model, LayoutLMv2ForTokenClassification, LayoutLMv2ForRelationExtraction +from paddlenlp.transformers import AutoModel -__all__ = ["LayoutXLMForSer", 'LayoutLMForSer'] +__all__ = ["LayoutXLMForSer", "LayoutLMForSer"] pretrained_model_dict = { - LayoutXLMModel: 'layoutxlm-base-uncased', - LayoutLMModel: 'layoutlm-base-uncased', - LayoutLMv2Model: 'layoutlmv2-base-uncased' + LayoutXLMModel: { + "base": "layoutxlm-base-uncased", + "vi": "layoutxlm-wo-backbone-base-uncased", + }, + LayoutLMModel: { + "base": "layoutlm-base-uncased", + }, + LayoutLMv2Model: { + "base": "layoutlmv2-base-uncased", + "vi": "layoutlmv2-wo-backbone-base-uncased", + }, } @@ -36,42 +45,47 @@ class NLPBaseModel(nn.Layer): def __init__(self, base_model_class, model_class, - type='ser', + mode="base", + type="ser", pretrained=True, checkpoints=None, **kwargs): super(NLPBaseModel, self).__init__() - if checkpoints is not None: + if checkpoints is not None: # load the trained model self.model = model_class.from_pretrained(checkpoints) - elif isinstance(pretrained, (str, )) and os.path.exists(pretrained): - self.model = model_class.from_pretrained(pretrained) - else: - pretrained_model_name = pretrained_model_dict[base_model_class] + else: # load the pretrained-model + pretrained_model_name = pretrained_model_dict[base_model_class][ + mode] if pretrained is True: base_model = base_model_class.from_pretrained( pretrained_model_name) else: - base_model = base_model_class( - **base_model_class.pretrained_init_configuration[ - pretrained_model_name]) - if type == 'ser': + base_model = base_model_class.from_pretrained(pretrained) + if type == "ser": self.model = model_class( - base_model, num_classes=kwargs['num_classes'], dropout=None) + base_model, num_classes=kwargs["num_classes"], dropout=None) else: self.model = model_class(base_model, dropout=None) self.out_channels = 1 + self.use_visual_backbone = True class LayoutLMForSer(NLPBaseModel): - def __init__(self, num_classes, pretrained=True, checkpoints=None, + def __init__(self, + num_classes, + pretrained=True, + checkpoints=None, + mode="base", **kwargs): super(LayoutLMForSer, self).__init__( LayoutLMModel, LayoutLMForTokenClassification, - 'ser', + mode, + "ser", pretrained, checkpoints, - num_classes=num_classes) + num_classes=num_classes, ) + self.use_visual_backbone = False def forward(self, x): x = self.model( @@ -85,62 +99,92 @@ class LayoutLMForSer(NLPBaseModel): class LayoutLMv2ForSer(NLPBaseModel): - def __init__(self, num_classes, pretrained=True, checkpoints=None, + def __init__(self, + num_classes, + pretrained=True, + checkpoints=None, + mode="base", **kwargs): super(LayoutLMv2ForSer, self).__init__( LayoutLMv2Model, LayoutLMv2ForTokenClassification, - 'ser', + mode, + "ser", pretrained, checkpoints, num_classes=num_classes) + self.use_visual_backbone = True + if hasattr(self.model.layoutlmv2, "use_visual_backbone" + ) and self.model.layoutlmv2.use_visual_backbone is False: + self.use_visual_backbone = False def forward(self, x): + if self.use_visual_backbone is True: + image = x[4] + else: + image = None x = self.model( input_ids=x[0], bbox=x[1], attention_mask=x[2], token_type_ids=x[3], - image=x[4], + image=image, position_ids=None, head_mask=None, labels=None) - if not self.training: + if self.training: + res = {"backbone_out": x[0]} + res.update(x[1]) + return res + else: return x - return x[0] class LayoutXLMForSer(NLPBaseModel): - def __init__(self, num_classes, pretrained=True, checkpoints=None, + def __init__(self, + num_classes, + pretrained=True, + checkpoints=None, + mode="base", **kwargs): super(LayoutXLMForSer, self).__init__( LayoutXLMModel, LayoutXLMForTokenClassification, - 'ser', + mode, + "ser", pretrained, checkpoints, num_classes=num_classes) + self.use_visual_backbone = True def forward(self, x): + if self.use_visual_backbone is True: + image = x[4] + else: + image = None x = self.model( input_ids=x[0], bbox=x[1], attention_mask=x[2], token_type_ids=x[3], - image=x[4], + image=image, position_ids=None, head_mask=None, labels=None) - if not self.training: + if self.training: + res = {"backbone_out": x[0]} + res.update(x[1]) + return res + else: return x - return x[0] class LayoutLMv2ForRe(NLPBaseModel): - def __init__(self, pretrained=True, checkpoints=None, **kwargs): - super(LayoutLMv2ForRe, self).__init__(LayoutLMv2Model, - LayoutLMv2ForRelationExtraction, - 're', pretrained, checkpoints) + def __init__(self, pretrained=True, checkpoints=None, mode="base", + **kwargs): + super(LayoutLMv2ForRe, self).__init__( + LayoutLMv2Model, LayoutLMv2ForRelationExtraction, mode, "re", + pretrained, checkpoints) def forward(self, x): x = self.model( @@ -158,18 +202,27 @@ class LayoutLMv2ForRe(NLPBaseModel): class LayoutXLMForRe(NLPBaseModel): - def __init__(self, pretrained=True, checkpoints=None, **kwargs): - super(LayoutXLMForRe, self).__init__(LayoutXLMModel, - LayoutXLMForRelationExtraction, - 're', pretrained, checkpoints) + def __init__(self, pretrained=True, checkpoints=None, mode="base", + **kwargs): + super(LayoutXLMForRe, self).__init__( + LayoutXLMModel, LayoutXLMForRelationExtraction, mode, "re", + pretrained, checkpoints) + self.use_visual_backbone = True + if hasattr(self.model.layoutxlm, "use_visual_backbone" + ) and self.model.layoutxlm.use_visual_backbone is False: + self.use_visual_backbone = False def forward(self, x): + if self.use_visual_backbone is True: + image = x[4] + else: + image = None x = self.model( input_ids=x[0], bbox=x[1], attention_mask=x[2], token_type_ids=x[3], - image=x[4], + image=image, position_ids=None, head_mask=None, labels=None, diff --git a/ppocr/postprocess/__init__.py b/ppocr/postprocess/__init__.py index eeebc5803f..6fa871a45c 100644 --- a/ppocr/postprocess/__init__.py +++ b/ppocr/postprocess/__init__.py @@ -31,21 +31,38 @@ from .rec_postprocess import CTCLabelDecode, AttnLabelDecode, SRNLabelDecode, \ SPINLabelDecode from .cls_postprocess import ClsPostProcess from .pg_postprocess import PGPostProcess -from .vqa_token_ser_layoutlm_postprocess import VQASerTokenLayoutLMPostProcess -from .vqa_token_re_layoutlm_postprocess import VQAReTokenLayoutLMPostProcess +from .vqa_token_ser_layoutlm_postprocess import VQASerTokenLayoutLMPostProcess, DistillationSerPostProcess +from .vqa_token_re_layoutlm_postprocess import VQAReTokenLayoutLMPostProcess, DistillationRePostProcess from .table_postprocess import TableMasterLabelDecode, TableLabelDecode def build_post_process(config, global_config=None): support_dict = [ - 'DBPostProcess', 'EASTPostProcess', 'SASTPostProcess', 'FCEPostProcess', - 'CTCLabelDecode', 'AttnLabelDecode', 'ClsPostProcess', 'SRNLabelDecode', - 'PGPostProcess', 'DistillationCTCLabelDecode', 'TableLabelDecode', - 'DistillationDBPostProcess', 'NRTRLabelDecode', 'SARLabelDecode', - 'SEEDLabelDecode', 'VQASerTokenLayoutLMPostProcess', - 'VQAReTokenLayoutLMPostProcess', 'PRENLabelDecode', - 'DistillationSARLabelDecode', 'ViTSTRLabelDecode', 'ABINetLabelDecode', - 'TableMasterLabelDecode', 'SPINLabelDecode' + 'DBPostProcess', + 'EASTPostProcess', + 'SASTPostProcess', + 'FCEPostProcess', + 'CTCLabelDecode', + 'AttnLabelDecode', + 'ClsPostProcess', + 'SRNLabelDecode', + 'PGPostProcess', + 'DistillationCTCLabelDecode', + 'TableLabelDecode', + 'DistillationDBPostProcess', + 'NRTRLabelDecode', + 'SARLabelDecode', + 'SEEDLabelDecode', + 'VQASerTokenLayoutLMPostProcess', + 'VQAReTokenLayoutLMPostProcess', + 'PRENLabelDecode', + 'DistillationSARLabelDecode', + 'ViTSTRLabelDecode', + 'ABINetLabelDecode', + 'TableMasterLabelDecode', + 'SPINLabelDecode', + 'DistillationSerPostProcess', + 'DistillationRePostProcess', ] if config['name'] == 'PSEPostProcess': diff --git a/ppocr/postprocess/vqa_token_re_layoutlm_postprocess.py b/ppocr/postprocess/vqa_token_re_layoutlm_postprocess.py index 1d55d13d76..96c25d9aac 100644 --- a/ppocr/postprocess/vqa_token_re_layoutlm_postprocess.py +++ b/ppocr/postprocess/vqa_token_re_layoutlm_postprocess.py @@ -49,3 +49,25 @@ class VQAReTokenLayoutLMPostProcess(object): result.append((ocr_info_head, ocr_info_tail)) results.append(result) return results + + +class DistillationRePostProcess(VQAReTokenLayoutLMPostProcess): + """ + DistillationRePostProcess + """ + + def __init__(self, model_name=["Student"], key=None, **kwargs): + super().__init__(**kwargs) + if not isinstance(model_name, list): + model_name = [model_name] + self.model_name = model_name + self.key = key + + def __call__(self, preds, *args, **kwargs): + output = dict() + for name in self.model_name: + pred = preds[name] + if self.key is not None: + pred = pred[self.key] + output[name] = super().__call__(pred, *args, **kwargs) + return output diff --git a/ppocr/postprocess/vqa_token_ser_layoutlm_postprocess.py b/ppocr/postprocess/vqa_token_ser_layoutlm_postprocess.py index 8a6669f71f..5541da90a0 100644 --- a/ppocr/postprocess/vqa_token_ser_layoutlm_postprocess.py +++ b/ppocr/postprocess/vqa_token_ser_layoutlm_postprocess.py @@ -93,3 +93,25 @@ class VQASerTokenLayoutLMPostProcess(object): ocr_info[idx]["pred"] = self.id2label_map_for_show[int(pred_id)] results.append(ocr_info) return results + + +class DistillationSerPostProcess(VQASerTokenLayoutLMPostProcess): + """ + DistillationSerPostProcess + """ + + def __init__(self, class_path, model_name=["Student"], key=None, **kwargs): + super().__init__(class_path, **kwargs) + if not isinstance(model_name, list): + model_name = [model_name] + self.model_name = model_name + self.key = key + + def __call__(self, preds, batch=None, *args, **kwargs): + output = dict() + for name in self.model_name: + pred = preds[name] + if self.key is not None: + pred = pred[self.key] + output[name] = super().__call__(pred, batch=batch, *args, **kwargs) + return output diff --git a/ppocr/utils/save_load.py b/ppocr/utils/save_load.py index 3647111fdd..8fded687c6 100644 --- a/ppocr/utils/save_load.py +++ b/ppocr/utils/save_load.py @@ -55,6 +55,9 @@ def load_model(config, model, optimizer=None, model_type='det'): best_model_dict = {} if model_type == 'vqa': + # NOTE: for vqa model, resume training is not supported now + if config["Architecture"]["algorithm"] in ["Distillation"]: + return best_model_dict checkpoints = config['Architecture']['Backbone']['checkpoints'] # load vqa method metric if checkpoints: @@ -78,6 +81,7 @@ def load_model(config, model, optimizer=None, model_type='det'): logger.warning( "{}.pdopt is not exists, params of optimizer is not loaded". format(checkpoints)) + return best_model_dict if checkpoints: @@ -166,15 +170,19 @@ def save_model(model, """ _mkdir_if_not_exist(model_path, logger) model_prefix = os.path.join(model_path, prefix) - paddle.save(optimizer.state_dict(), model_prefix + '.pdopt') + if config['Architecture']["model_type"] != 'vqa': + paddle.save(optimizer.state_dict(), model_prefix + '.pdopt') if config['Architecture']["model_type"] != 'vqa': paddle.save(model.state_dict(), model_prefix + '.pdparams') metric_prefix = model_prefix - else: + else: # for vqa system, we follow the save/load rules in NLP if config['Global']['distributed']: - model._layers.backbone.model.save_pretrained(model_prefix) + arch = model._layers else: - model.backbone.model.save_pretrained(model_prefix) + arch = model + if config["Architecture"]["algorithm"] in ["Distillation"]: + arch = arch.Student + arch.backbone.model.save_pretrained(model_prefix) metric_prefix = os.path.join(model_prefix, 'metric') # save metric and config with open(metric_prefix + '.states', 'wb') as f: diff --git a/tools/infer/utility.py b/tools/infer/utility.py index 7eb77dec74..9345106e77 100644 --- a/tools/infer/utility.py +++ b/tools/infer/utility.py @@ -38,6 +38,7 @@ def init_args(): parser.add_argument("--ir_optim", type=str2bool, default=True) parser.add_argument("--use_tensorrt", type=str2bool, default=False) parser.add_argument("--min_subgraph_size", type=int, default=15) + parser.add_argument("--shape_info_filename", type=str, default=None) parser.add_argument("--precision", type=str, default="fp32") parser.add_argument("--gpu_mem", type=int, default=500) @@ -204,9 +205,18 @@ def create_predictor(args, mode, logger): workspace_size=1 << 30, precision_mode=precision, max_batch_size=args.max_batch_size, - min_subgraph_size=args.min_subgraph_size, + min_subgraph_size=args.min_subgraph_size, # skip the minmum trt subgraph use_calib_mode=False) - # skip the minmum trt subgraph + + # collect shape + if args.shape_info_filename is not None: + if not os.path.exists(args.shape_info_filename): + config.collect_shape_range_info(args.shape_info_filename) + logger.info(f"collect dynamic shape info into : {args.shape_info_filename}") + else: + logger.info(f"dynamic shape info file( {args.shape_info_filename} ) already exists, not need to generate again.") + config.enable_tuned_tensorrt_dynamic_shape(args.shape_info_filename, True) + use_dynamic_shape = True if mode == "det": min_input_shape = { diff --git a/tools/infer_vqa_token_ser_re.py b/tools/infer_vqa_token_ser_re.py index 20ab1fe176..51378bdaeb 100755 --- a/tools/infer_vqa_token_ser_re.py +++ b/tools/infer_vqa_token_ser_re.py @@ -113,10 +113,13 @@ def make_input(ser_inputs, ser_results): class SerRePredictor(object): def __init__(self, config, ser_config): + global_config = config['Global'] + if "infer_mode" in global_config: + ser_config["Global"]["infer_mode"] = global_config["infer_mode"] + self.ser_engine = SerPredictor(ser_config) # init re model - global_config = config['Global'] # build post process self.post_process_class = build_post_process(config['PostProcess'], @@ -130,8 +133,8 @@ class SerRePredictor(object): self.model.eval() - def __call__(self, img_path): - ser_results, ser_inputs = self.ser_engine({'img_path': img_path}) + def __call__(self, data): + ser_results, ser_inputs = self.ser_engine(data) re_input, entity_idx_dict_batch = make_input(ser_inputs, ser_results) preds = self.model(re_input) post_result = self.post_process_class( @@ -173,18 +176,33 @@ if __name__ == '__main__': ser_re_engine = SerRePredictor(config, ser_config) - infer_imgs = get_image_file_list(config['Global']['infer_img']) + if config["Global"].get("infer_mode", None) is False: + data_dir = config['Eval']['dataset']['data_dir'] + with open(config['Global']['infer_img'], "rb") as f: + infer_imgs = f.readlines() + else: + infer_imgs = get_image_file_list(config['Global']['infer_img']) + with open( os.path.join(config['Global']['save_res_path'], "infer_results.txt"), "w", encoding='utf-8') as fout: - for idx, img_path in enumerate(infer_imgs): + for idx, info in enumerate(infer_imgs): + if config["Global"].get("infer_mode", None) is False: + data_line = info.decode('utf-8') + substr = data_line.strip("\n").split("\t") + img_path = os.path.join(data_dir, substr[0]) + data = {'img_path': img_path, 'label': substr[1]} + else: + img_path = info + data = {'img_path': img_path} + save_img_path = os.path.join( config['Global']['save_res_path'], os.path.splitext(os.path.basename(img_path))[0] + "_ser_re.jpg") - result = ser_re_engine(img_path) + result = ser_re_engine(data) result = result[0] fout.write(img_path + "\t" + json.dumps( { From 03d384860fce041f1b18633c8969c097ba6537d2 Mon Sep 17 00:00:00 2001 From: andyjpaddle Date: Mon, 8 Aug 2022 08:58:07 +0000 Subject: [PATCH 23/25] fix amp train for re --- tools/infer/predict_det_eval.py | 363 ++++++++++++++++++++++ tools/infer/predict_rec_eval.py | 534 ++++++++++++++++++++++++++++++++ tools/program.py | 8 +- 3 files changed, 902 insertions(+), 3 deletions(-) create mode 100755 tools/infer/predict_det_eval.py create mode 100755 tools/infer/predict_rec_eval.py diff --git a/tools/infer/predict_det_eval.py b/tools/infer/predict_det_eval.py new file mode 100755 index 0000000000..d1f832036b --- /dev/null +++ b/tools/infer/predict_det_eval.py @@ -0,0 +1,363 @@ +# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +import os +import sys + +__dir__ = os.path.dirname(os.path.abspath(__file__)) +sys.path.append(__dir__) +sys.path.insert(0, os.path.abspath(os.path.join(__dir__, '../..'))) + +os.environ["FLAGS_allocator_strategy"] = 'auto_growth' + +import cv2 +import numpy as np +import time +import sys + +import tools.infer.utility as utility +from ppocr.utils.logging import get_logger +from ppocr.utils.utility import get_image_file_list, check_and_read_gif +from ppocr.data import create_operators, transform +from ppocr.postprocess import build_post_process +import json +logger = get_logger() + + +class TextDetector(object): + def __init__(self, args): + self.args = args + self.det_algorithm = args.det_algorithm + self.use_onnx = args.use_onnx + pre_process_list = [{ + 'DetResizeForTest': { + 'limit_side_len': args.det_limit_side_len, + 'limit_type': args.det_limit_type, + } + }, { + 'NormalizeImage': { + 'std': [0.229, 0.224, 0.225], + 'mean': [0.485, 0.456, 0.406], + 'scale': '1./255.', + 'order': 'hwc' + } + }, { + 'ToCHWImage': None + }, { + 'KeepKeys': { + 'keep_keys': ['image', 'shape'] + } + }] + postprocess_params = {} + if self.det_algorithm == "DB": + postprocess_params['name'] = 'DBPostProcess' + postprocess_params["thresh"] = args.det_db_thresh + postprocess_params["box_thresh"] = args.det_db_box_thresh + postprocess_params["max_candidates"] = 1000 + postprocess_params["unclip_ratio"] = args.det_db_unclip_ratio + postprocess_params["use_dilation"] = args.use_dilation + postprocess_params["score_mode"] = args.det_db_score_mode + elif self.det_algorithm == "EAST": + postprocess_params['name'] = 'EASTPostProcess' + postprocess_params["score_thresh"] = args.det_east_score_thresh + postprocess_params["cover_thresh"] = args.det_east_cover_thresh + postprocess_params["nms_thresh"] = args.det_east_nms_thresh + elif self.det_algorithm == "SAST": + pre_process_list[0] = { + 'DetResizeForTest': { + 'resize_long': args.det_limit_side_len + } + } + postprocess_params['name'] = 'SASTPostProcess' + postprocess_params["score_thresh"] = args.det_sast_score_thresh + postprocess_params["nms_thresh"] = args.det_sast_nms_thresh + self.det_sast_polygon = args.det_sast_polygon + if self.det_sast_polygon: + postprocess_params["sample_pts_num"] = 6 + postprocess_params["expand_scale"] = 1.2 + postprocess_params["shrink_ratio_of_width"] = 0.2 + else: + postprocess_params["sample_pts_num"] = 2 + postprocess_params["expand_scale"] = 1.0 + postprocess_params["shrink_ratio_of_width"] = 0.3 + elif self.det_algorithm == "PSE": + postprocess_params['name'] = 'PSEPostProcess' + postprocess_params["thresh"] = args.det_pse_thresh + postprocess_params["box_thresh"] = args.det_pse_box_thresh + postprocess_params["min_area"] = args.det_pse_min_area + postprocess_params["box_type"] = args.det_pse_box_type + postprocess_params["scale"] = args.det_pse_scale + self.det_pse_box_type = args.det_pse_box_type + elif self.det_algorithm == "FCE": + pre_process_list[0] = { + 'DetResizeForTest': { + 'rescale_img': [1080, 736] + } + } + postprocess_params['name'] = 'FCEPostProcess' + postprocess_params["scales"] = args.scales + postprocess_params["alpha"] = args.alpha + postprocess_params["beta"] = args.beta + postprocess_params["fourier_degree"] = args.fourier_degree + postprocess_params["box_type"] = args.det_fce_box_type + else: + logger.info("unknown det_algorithm:{}".format(self.det_algorithm)) + sys.exit(0) + + self.preprocess_op = create_operators(pre_process_list) + self.postprocess_op = build_post_process(postprocess_params) + self.predictor, self.input_tensor, self.output_tensors, self.config = utility.create_predictor( + args, 'det', logger) + + if self.use_onnx: + img_h, img_w = self.input_tensor.shape[2:] + if img_h is not None and img_w is not None and img_h > 0 and img_w > 0: + pre_process_list[0] = { + 'DetResizeForTest': { + 'image_shape': [img_h, img_w] + } + } + self.preprocess_op = create_operators(pre_process_list) + + if args.benchmark: + import auto_log + pid = os.getpid() + gpu_id = utility.get_infer_gpuid() + self.autolog = auto_log.AutoLogger( + model_name="det", + model_precision=args.precision, + batch_size=1, + data_shape="dynamic", + save_path=None, + inference_config=self.config, + pids=pid, + process_name=None, + gpu_ids=gpu_id if args.use_gpu else None, + time_keys=[ + 'preprocess_time', 'inference_time', 'postprocess_time' + ], + warmup=2, + logger=logger) + + def order_points_clockwise(self, pts): + rect = np.zeros((4, 2), dtype="float32") + s = pts.sum(axis=1) + rect[0] = pts[np.argmin(s)] + rect[2] = pts[np.argmax(s)] + tmp = np.delete(pts, (np.argmin(s), np.argmax(s)), axis=0) + diff = np.diff(np.array(tmp), axis=1) + rect[1] = tmp[np.argmin(diff)] + rect[3] = tmp[np.argmax(diff)] + return rect + + def clip_det_res(self, points, img_height, img_width): + for pno in range(points.shape[0]): + points[pno, 0] = int(min(max(points[pno, 0], 0), img_width - 1)) + points[pno, 1] = int(min(max(points[pno, 1], 0), img_height - 1)) + return points + + def filter_tag_det_res(self, dt_boxes, image_shape): + img_height, img_width = image_shape[0:2] + dt_boxes_new = [] + for box in dt_boxes: + box = self.order_points_clockwise(box) + box = self.clip_det_res(box, img_height, img_width) + rect_width = int(np.linalg.norm(box[0] - box[1])) + rect_height = int(np.linalg.norm(box[0] - box[3])) + if rect_width <= 3 or rect_height <= 3: + continue + dt_boxes_new.append(box) + dt_boxes = np.array(dt_boxes_new) + return dt_boxes + + def filter_tag_det_res_only_clip(self, dt_boxes, image_shape): + img_height, img_width = image_shape[0:2] + dt_boxes_new = [] + for box in dt_boxes: + box = self.clip_det_res(box, img_height, img_width) + dt_boxes_new.append(box) + dt_boxes = np.array(dt_boxes_new) + return dt_boxes + + def __call__(self, img): + ori_im = img.copy() + data = {'image': img} + + st = time.time() + + if self.args.benchmark: + self.autolog.times.start() + + data = transform(data, self.preprocess_op) + img, shape_list = data + if img is None: + return None, 0 + img = np.expand_dims(img, axis=0) + shape_list = np.expand_dims(shape_list, axis=0) + img = img.copy() + + if self.args.benchmark: + self.autolog.times.stamp() + if self.use_onnx: + input_dict = {} + input_dict[self.input_tensor.name] = img + outputs = self.predictor.run(self.output_tensors, input_dict) + else: + self.input_tensor.copy_from_cpu(img) + self.predictor.run() + outputs = [] + for output_tensor in self.output_tensors: + output = output_tensor.copy_to_cpu() + outputs.append(output) + if self.args.benchmark: + self.autolog.times.stamp() + + preds = {} + if self.det_algorithm == "EAST": + preds['f_geo'] = outputs[0] + preds['f_score'] = outputs[1] + elif self.det_algorithm == 'SAST': + preds['f_border'] = outputs[0] + preds['f_score'] = outputs[1] + preds['f_tco'] = outputs[2] + preds['f_tvo'] = outputs[3] + elif self.det_algorithm in ['DB', 'PSE']: + preds['maps'] = outputs[0] + elif self.det_algorithm == 'FCE': + for i, output in enumerate(outputs): + preds['level_{}'.format(i)] = output + else: + raise NotImplementedError + + #self.predictor.try_shrink_memory() + post_result = self.postprocess_op(preds, shape_list) + dt_boxes = post_result[0]['points'] + if (self.det_algorithm == "SAST" and self.det_sast_polygon) or ( + self.det_algorithm in ["PSE", "FCE"] and + self.postprocess_op.box_type == 'poly'): + dt_boxes = self.filter_tag_det_res_only_clip(dt_boxes, ori_im.shape) + else: + dt_boxes = self.filter_tag_det_res(dt_boxes, ori_im.shape) + + if self.args.benchmark: + self.autolog.times.end(stamp=True) + et = time.time() + return dt_boxes, et - st + + +if __name__ == "__main__": + from ppocr.metrics.eval_det_iou import DetectionIoUEvaluator + evaluator = DetectionIoUEvaluator() + args = utility.parse_args() + + # image_file_list = get_image_file_list(args.image_dir) + def _check_image_file(path): + img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif'} + return any([path.lower().endswith(e) for e in img_end]) + + def get_image_file_list_from_txt(img_file): + imgs_lists = [] + label_lists = [] + if img_file is None or not os.path.exists(img_file): + raise Exception("not found any img file in {}".format(img_file)) + + img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif'} + root_dir = img_file.split('/')[0] + with open(img_file, 'r') as f: + lines = f.readlines() + for line in lines: + line = line.replace('\n', '').split('\t') + file_path, label = line[0], line[1] + file_path = os.path.join(root_dir, file_path) + if os.path.isfile(file_path) and _check_image_file(file_path): + imgs_lists.append(file_path) + label_lists.append(label) + + if len(imgs_lists) == 0: + raise Exception("not found any img file in {}".format(img_file)) + return imgs_lists, label_lists + + image_file_list, label_list = get_image_file_list_from_txt(args.image_dir) + + text_detector = TextDetector(args) + count = 0 + total_time = 0 + draw_img_save = "./inference_results" + + if args.warmup: + img = np.random.uniform(0, 255, [640, 640, 3]).astype(np.uint8) + for i in range(2): + res = text_detector(img) + + if not os.path.exists(draw_img_save): + os.makedirs(draw_img_save) + save_results = [] + results = [] + for idx in range(len(image_file_list)): + image_file = image_file_list[idx] + label = json.loads(label_list[idx]) + img, flag = check_and_read_gif(image_file) + if not flag: + img = cv2.imread(image_file) + if img is None: + logger.info("error in loading image:{}".format(image_file)) + continue + st = time.time() + dt_boxes, _ = text_detector(img) + elapse = time.time() - st + if count > 0: + total_time += elapse + count += 1 + save_pred = os.path.basename(image_file) + "\t" + str( + json.dumps([x.tolist() for x in dt_boxes])) + "\n" + save_results.append(save_pred) + + # for eval + gt_info_list = [] + det_info_list = [] + for dt_box in dt_boxes: + det_info = { + 'points': np.array( + dt_box, dtype=np.float32), + 'text': '' + } + det_info_list.append(det_info) + for lab in label: + gt_info = { + 'points': np.array( + lab['points'], dtype=np.float32), + 'text': '', + 'ignore': False + } + gt_info_list.append(gt_info) + result = evaluator.evaluate_image(gt_info_list, det_info_list) + results.append(result) + + metrics = evaluator.combine_results(results) + print('predict det eval on ', args.image_dir) + print('metrics: ', metrics) + +# logger.info(save_pred) +# logger.info("The predict time of {}: {}".format(image_file, elapse)) +# src_im = utility.draw_text_det_res(dt_boxes, image_file) +# img_name_pure = os.path.split(image_file)[-1] +# img_path = os.path.join(draw_img_save, +# "det_res_{}".format(img_name_pure)) +# cv2.imwrite(img_path, src_im) +# logger.info("The visualized image saved in {}".format(img_path)) + +# with open(os.path.join(draw_img_save, "det_results.txt"), 'w') as f: +# f.writelines(save_results) +# f.close() +# if args.benchmark: +# text_detector.autolog.report() diff --git a/tools/infer/predict_rec_eval.py b/tools/infer/predict_rec_eval.py new file mode 100755 index 0000000000..3150d11ddf --- /dev/null +++ b/tools/infer/predict_rec_eval.py @@ -0,0 +1,534 @@ +# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +import os +import sys +from PIL import Image +__dir__ = os.path.dirname(os.path.abspath(__file__)) +sys.path.append(__dir__) +sys.path.insert(0, os.path.abspath(os.path.join(__dir__, '../..'))) + +os.environ["FLAGS_allocator_strategy"] = 'auto_growth' + +import cv2 +import numpy as np +import math +import time +import traceback +import paddle + +import tools.infer.utility as utility +from ppocr.postprocess import build_post_process +from ppocr.utils.logging import get_logger +from ppocr.utils.utility import get_image_file_list, check_and_read_gif + +logger = get_logger() + + +class TextRecognizer(object): + def __init__(self, args): + self.rec_image_shape = [int(v) for v in args.rec_image_shape.split(",")] + self.rec_batch_num = args.rec_batch_num + self.rec_algorithm = args.rec_algorithm + postprocess_params = { + 'name': 'CTCLabelDecode', + "character_dict_path": args.rec_char_dict_path, + "use_space_char": args.use_space_char + } + if self.rec_algorithm == "SRN": + postprocess_params = { + 'name': 'SRNLabelDecode', + "character_dict_path": args.rec_char_dict_path, + "use_space_char": args.use_space_char + } + elif self.rec_algorithm == "RARE": + postprocess_params = { + 'name': 'AttnLabelDecode', + "character_dict_path": args.rec_char_dict_path, + "use_space_char": args.use_space_char + } + elif self.rec_algorithm == 'NRTR': + postprocess_params = { + 'name': 'NRTRLabelDecode', + "character_dict_path": args.rec_char_dict_path, + "use_space_char": args.use_space_char + } + elif self.rec_algorithm == "SAR": + postprocess_params = { + 'name': 'SARLabelDecode', + "character_dict_path": args.rec_char_dict_path, + "use_space_char": args.use_space_char + } + elif self.rec_algorithm == 'ViTSTR': + postprocess_params = { + 'name': 'ViTSTRLabelDecode', + "character_dict_path": args.rec_char_dict_path, + "use_space_char": args.use_space_char + } + elif self.rec_algorithm == 'ABINet': + postprocess_params = { + 'name': 'ABINetLabelDecode', + "character_dict_path": args.rec_char_dict_path, + "use_space_char": args.use_space_char + } + self.postprocess_op = build_post_process(postprocess_params) + self.predictor, self.input_tensor, self.output_tensors, self.config = \ + utility.create_predictor(args, 'rec', logger) + self.benchmark = args.benchmark + self.use_onnx = args.use_onnx + if args.benchmark: + import auto_log + pid = os.getpid() + gpu_id = utility.get_infer_gpuid() + self.autolog = auto_log.AutoLogger( + model_name="rec", + model_precision=args.precision, + batch_size=args.rec_batch_num, + data_shape="dynamic", + save_path=None, #args.save_log_path, + inference_config=self.config, + pids=pid, + process_name=None, + gpu_ids=gpu_id if args.use_gpu else None, + time_keys=[ + 'preprocess_time', 'inference_time', 'postprocess_time' + ], + warmup=0, + logger=logger) + + def resize_norm_img(self, img, max_wh_ratio): + imgC, imgH, imgW = self.rec_image_shape + if self.rec_algorithm == 'NRTR' or self.rec_algorithm == 'ViTSTR': + img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) + # return padding_im + image_pil = Image.fromarray(np.uint8(img)) + if self.rec_algorithm == 'ViTSTR': + img = image_pil.resize([imgW, imgH], Image.BICUBIC) + else: + img = image_pil.resize([imgW, imgH], Image.ANTIALIAS) + img = np.array(img) + norm_img = np.expand_dims(img, -1) + norm_img = norm_img.transpose((2, 0, 1)) + if self.rec_algorithm == 'ViTSTR': + norm_img = norm_img.astype(np.float32) / 255. + else: + norm_img = norm_img.astype(np.float32) / 128. - 1. + return norm_img + + assert imgC == img.shape[2] + imgW = int((imgH * max_wh_ratio)) + if self.use_onnx: + w = self.input_tensor.shape[3:][0] + if w is not None and w > 0: + imgW = w + + h, w = img.shape[:2] + ratio = w / float(h) + if math.ceil(imgH * ratio) > imgW: + resized_w = imgW + else: + resized_w = int(math.ceil(imgH * ratio)) + if self.rec_algorithm == 'RARE': + if resized_w > self.rec_image_shape[2]: + resized_w = self.rec_image_shape[2] + imgW = self.rec_image_shape[2] + resized_image = cv2.resize(img, (resized_w, imgH)) + resized_image = resized_image.astype('float32') + resized_image = resized_image.transpose((2, 0, 1)) / 255 + resized_image -= 0.5 + resized_image /= 0.5 + padding_im = np.zeros((imgC, imgH, imgW), dtype=np.float32) + padding_im[:, :, 0:resized_w] = resized_image + return padding_im + + def resize_norm_img_srn(self, img, image_shape): + imgC, imgH, imgW = image_shape + + img_black = np.zeros((imgH, imgW)) + im_hei = img.shape[0] + im_wid = img.shape[1] + + if im_wid <= im_hei * 1: + img_new = cv2.resize(img, (imgH * 1, imgH)) + elif im_wid <= im_hei * 2: + img_new = cv2.resize(img, (imgH * 2, imgH)) + elif im_wid <= im_hei * 3: + img_new = cv2.resize(img, (imgH * 3, imgH)) + else: + img_new = cv2.resize(img, (imgW, imgH)) + + img_np = np.asarray(img_new) + img_np = cv2.cvtColor(img_np, cv2.COLOR_BGR2GRAY) + img_black[:, 0:img_np.shape[1]] = img_np + img_black = img_black[:, :, np.newaxis] + + row, col, c = img_black.shape + c = 1 + + return np.reshape(img_black, (c, row, col)).astype(np.float32) + + def srn_other_inputs(self, image_shape, num_heads, max_text_length): + + imgC, imgH, imgW = image_shape + feature_dim = int((imgH / 8) * (imgW / 8)) + + encoder_word_pos = np.array(range(0, feature_dim)).reshape( + (feature_dim, 1)).astype('int64') + gsrm_word_pos = np.array(range(0, max_text_length)).reshape( + (max_text_length, 1)).astype('int64') + + gsrm_attn_bias_data = np.ones((1, max_text_length, max_text_length)) + gsrm_slf_attn_bias1 = np.triu(gsrm_attn_bias_data, 1).reshape( + [-1, 1, max_text_length, max_text_length]) + gsrm_slf_attn_bias1 = np.tile( + gsrm_slf_attn_bias1, + [1, num_heads, 1, 1]).astype('float32') * [-1e9] + + gsrm_slf_attn_bias2 = np.tril(gsrm_attn_bias_data, -1).reshape( + [-1, 1, max_text_length, max_text_length]) + gsrm_slf_attn_bias2 = np.tile( + gsrm_slf_attn_bias2, + [1, num_heads, 1, 1]).astype('float32') * [-1e9] + + encoder_word_pos = encoder_word_pos[np.newaxis, :] + gsrm_word_pos = gsrm_word_pos[np.newaxis, :] + + return [ + encoder_word_pos, gsrm_word_pos, gsrm_slf_attn_bias1, + gsrm_slf_attn_bias2 + ] + + def process_image_srn(self, img, image_shape, num_heads, max_text_length): + norm_img = self.resize_norm_img_srn(img, image_shape) + norm_img = norm_img[np.newaxis, :] + + [encoder_word_pos, gsrm_word_pos, gsrm_slf_attn_bias1, gsrm_slf_attn_bias2] = \ + self.srn_other_inputs(image_shape, num_heads, max_text_length) + + gsrm_slf_attn_bias1 = gsrm_slf_attn_bias1.astype(np.float32) + gsrm_slf_attn_bias2 = gsrm_slf_attn_bias2.astype(np.float32) + encoder_word_pos = encoder_word_pos.astype(np.int64) + gsrm_word_pos = gsrm_word_pos.astype(np.int64) + + return (norm_img, encoder_word_pos, gsrm_word_pos, gsrm_slf_attn_bias1, + gsrm_slf_attn_bias2) + + def resize_norm_img_sar(self, img, image_shape, + width_downsample_ratio=0.25): + imgC, imgH, imgW_min, imgW_max = image_shape + h = img.shape[0] + w = img.shape[1] + valid_ratio = 1.0 + # make sure new_width is an integral multiple of width_divisor. + width_divisor = int(1 / width_downsample_ratio) + # resize + ratio = w / float(h) + resize_w = math.ceil(imgH * ratio) + if resize_w % width_divisor != 0: + resize_w = round(resize_w / width_divisor) * width_divisor + if imgW_min is not None: + resize_w = max(imgW_min, resize_w) + if imgW_max is not None: + valid_ratio = min(1.0, 1.0 * resize_w / imgW_max) + resize_w = min(imgW_max, resize_w) + resized_image = cv2.resize(img, (resize_w, imgH)) + resized_image = resized_image.astype('float32') + # norm + if image_shape[0] == 1: + resized_image = resized_image / 255 + resized_image = resized_image[np.newaxis, :] + else: + resized_image = resized_image.transpose((2, 0, 1)) / 255 + resized_image -= 0.5 + resized_image /= 0.5 + resize_shape = resized_image.shape + padding_im = -1.0 * np.ones((imgC, imgH, imgW_max), dtype=np.float32) + padding_im[:, :, 0:resize_w] = resized_image + pad_shape = padding_im.shape + + return padding_im, resize_shape, pad_shape, valid_ratio + + def resize_norm_img_svtr(self, img, image_shape): + + imgC, imgH, imgW = image_shape + resized_image = cv2.resize( + img, (imgW, imgH), interpolation=cv2.INTER_LINEAR) + resized_image = resized_image.astype('float32') + resized_image = resized_image.transpose((2, 0, 1)) / 255 + resized_image -= 0.5 + resized_image /= 0.5 + return resized_image + + def resize_norm_img_abinet(self, img, image_shape): + + imgC, imgH, imgW = image_shape + + resized_image = cv2.resize( + img, (imgW, imgH), interpolation=cv2.INTER_LINEAR) + resized_image = resized_image.astype('float32') + resized_image = resized_image / 255. + + mean = np.array([0.485, 0.456, 0.406]) + std = np.array([0.229, 0.224, 0.225]) + resized_image = ( + resized_image - mean[None, None, ...]) / std[None, None, ...] + resized_image = resized_image.transpose((2, 0, 1)) + resized_image = resized_image.astype('float32') + + return resized_image + + def __call__(self, img_list): + img_num = len(img_list) + # Calculate the aspect ratio of all text bars + width_list = [] + for img in img_list: + width_list.append(img.shape[1] / float(img.shape[0])) + # Sorting can speed up the recognition process + indices = np.argsort(np.array(width_list)) + rec_res = [['', 0.0]] * img_num + batch_num = self.rec_batch_num + st = time.time() + if self.benchmark: + self.autolog.times.start() + for beg_img_no in range(0, img_num, batch_num): + end_img_no = min(img_num, beg_img_no + batch_num) + norm_img_batch = [] + imgC, imgH, imgW = self.rec_image_shape + max_wh_ratio = imgW / imgH + # max_wh_ratio = 0 + for ino in range(beg_img_no, end_img_no): + h, w = img_list[indices[ino]].shape[0:2] + wh_ratio = w * 1.0 / h + max_wh_ratio = max(max_wh_ratio, wh_ratio) + for ino in range(beg_img_no, end_img_no): + + if self.rec_algorithm == "SAR": + norm_img, _, _, valid_ratio = self.resize_norm_img_sar( + img_list[indices[ino]], self.rec_image_shape) + norm_img = norm_img[np.newaxis, :] + valid_ratio = np.expand_dims(valid_ratio, axis=0) + valid_ratios = [] + valid_ratios.append(valid_ratio) + norm_img_batch.append(norm_img) + elif self.rec_algorithm == "SRN": + norm_img = self.process_image_srn( + img_list[indices[ino]], self.rec_image_shape, 8, 25) + encoder_word_pos_list = [] + gsrm_word_pos_list = [] + gsrm_slf_attn_bias1_list = [] + gsrm_slf_attn_bias2_list = [] + encoder_word_pos_list.append(norm_img[1]) + gsrm_word_pos_list.append(norm_img[2]) + gsrm_slf_attn_bias1_list.append(norm_img[3]) + gsrm_slf_attn_bias2_list.append(norm_img[4]) + norm_img_batch.append(norm_img[0]) + elif self.rec_algorithm == "SVTR": + norm_img = self.resize_norm_img_svtr(img_list[indices[ino]], + self.rec_image_shape) + norm_img = norm_img[np.newaxis, :] + norm_img_batch.append(norm_img) + elif self.rec_algorithm == "ABINet": + norm_img = self.resize_norm_img_abinet( + img_list[indices[ino]], self.rec_image_shape) + norm_img = norm_img[np.newaxis, :] + norm_img_batch.append(norm_img) + else: + norm_img = self.resize_norm_img(img_list[indices[ino]], + max_wh_ratio) + norm_img = norm_img[np.newaxis, :] + norm_img_batch.append(norm_img) + norm_img_batch = np.concatenate(norm_img_batch) + norm_img_batch = norm_img_batch.copy() + if self.benchmark: + self.autolog.times.stamp() + + if self.rec_algorithm == "SRN": + encoder_word_pos_list = np.concatenate(encoder_word_pos_list) + gsrm_word_pos_list = np.concatenate(gsrm_word_pos_list) + gsrm_slf_attn_bias1_list = np.concatenate( + gsrm_slf_attn_bias1_list) + gsrm_slf_attn_bias2_list = np.concatenate( + gsrm_slf_attn_bias2_list) + + inputs = [ + norm_img_batch, + encoder_word_pos_list, + gsrm_word_pos_list, + gsrm_slf_attn_bias1_list, + gsrm_slf_attn_bias2_list, + ] + if self.use_onnx: + input_dict = {} + input_dict[self.input_tensor.name] = norm_img_batch + outputs = self.predictor.run(self.output_tensors, + input_dict) + preds = {"predict": outputs[2]} + else: + input_names = self.predictor.get_input_names() + for i in range(len(input_names)): + input_tensor = self.predictor.get_input_handle( + input_names[i]) + input_tensor.copy_from_cpu(inputs[i]) + self.predictor.run() + outputs = [] + for output_tensor in self.output_tensors: + output = output_tensor.copy_to_cpu() + outputs.append(output) + if self.benchmark: + self.autolog.times.stamp() + preds = {"predict": outputs[2]} + elif self.rec_algorithm == "SAR": + valid_ratios = np.concatenate(valid_ratios) + inputs = [ + norm_img_batch, + valid_ratios, + ] + if self.use_onnx: + input_dict = {} + input_dict[self.input_tensor.name] = norm_img_batch + outputs = self.predictor.run(self.output_tensors, + input_dict) + preds = outputs[0] + else: + input_names = self.predictor.get_input_names() + for i in range(len(input_names)): + input_tensor = self.predictor.get_input_handle( + input_names[i]) + input_tensor.copy_from_cpu(inputs[i]) + self.predictor.run() + outputs = [] + for output_tensor in self.output_tensors: + output = output_tensor.copy_to_cpu() + outputs.append(output) + if self.benchmark: + self.autolog.times.stamp() + preds = outputs[0] + else: + if self.use_onnx: + input_dict = {} + input_dict[self.input_tensor.name] = norm_img_batch + outputs = self.predictor.run(self.output_tensors, + input_dict) + preds = outputs[0] + else: + self.input_tensor.copy_from_cpu(norm_img_batch) + self.predictor.run() + outputs = [] + for output_tensor in self.output_tensors: + output = output_tensor.copy_to_cpu() + outputs.append(output) + if self.benchmark: + self.autolog.times.stamp() + if len(outputs) != 1: + preds = outputs + else: + preds = outputs[0] + rec_result = self.postprocess_op(preds) + for rno in range(len(rec_result)): + rec_res[indices[beg_img_no + rno]] = rec_result[rno] + if self.benchmark: + self.autolog.times.end(stamp=True) + return rec_res, time.time() - st + + +def main(args): + # image_file_list = get_image_file_list(args.image_dir) + + def _check_image_file(path): + img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif'} + return any([path.lower().endswith(e) for e in img_end]) + + def get_image_file_list_from_txt(img_file): + imgs_lists = [] + label_lists = [] + if img_file is None or not os.path.exists(img_file): + raise Exception("not found any img file in {}".format(img_file)) + + img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif'} + root_dir = img_file.split('/')[0] + with open(img_file, 'r') as f: + lines = f.readlines() + for line in lines: + line = line.replace('\n', '').split('\t') + file_path, label = line[0], line[1] + file_path = os.path.join(root_dir, file_path) + if os.path.isfile(file_path) and _check_image_file(file_path): + imgs_lists.append(file_path) + label_lists.append(label) + + if len(imgs_lists) == 0: + raise Exception("not found any img file in {}".format(img_file)) + return imgs_lists, label_lists + + image_file_list, label_list = get_image_file_list_from_txt(args.image_dir) + + text_recognizer = TextRecognizer(args) + valid_image_file_list = [] + img_list = [] + + logger.info( + "In PP-OCRv3, rec_image_shape parameter defaults to '3, 48, 320', " + "if you are using recognition model with PP-OCRv2 or an older version, please set --rec_image_shape='3,32,320" + ) + # warmup 2 times + if args.warmup: + img = np.random.uniform(0, 255, [48, 320, 3]).astype(np.uint8) + for i in range(2): + res = text_recognizer([img] * int(args.rec_batch_num)) + + for image_file in image_file_list: + img, flag = check_and_read_gif(image_file) + if not flag: + img = cv2.imread(image_file) + if img is None: + logger.info("error in loading image:{}".format(image_file)) + continue + valid_image_file_list.append(image_file) + img_list.append(img) + + try: + rec_res, _ = text_recognizer(img_list) + except Exception as E: + logger.info(traceback.format_exc()) + logger.info(E) + exit() + correct_num = 0 + for ino in range(len(img_list)): + pred = rec_res[ino][0] + gt = label_list[ino] + if pred == gt: + correct_num += 1 + acc = correct_num * 1.0 / len(img_list) + print('predict rec eval on ', args.image_dir) + print('acc: ', acc) + + # for debug bad case + bad_case_lines = [] + for ino in range(len(img_list)): + pred = rec_res[ino][0] + gt = label_list[ino] + if pred != gt and len(gt) <= 25: + bad_case = valid_image_file_list[ + ino] + '\t' + 'pred:' + pred + '\t' + 'gt:' + gt + '\n' + bad_case_lines.append(bad_case) + + with open('bad_case_hwdb2.txt', 'a+') as f: + f.writelines(bad_case_lines) + # end debug case + + if args.benchmark: + text_recognizer.autolog.report() + + +if __name__ == "__main__": + main(utility.parse_args()) diff --git a/tools/program.py b/tools/program.py index 0fa0e609bd..dd8037e473 100755 --- a/tools/program.py +++ b/tools/program.py @@ -154,13 +154,14 @@ def check_xpu(use_xpu): except Exception as e: pass + def to_float32(preds): if isinstance(preds, dict): for k in preds: if isinstance(preds[k], dict) or isinstance(preds[k], list): preds[k] = to_float32(preds[k]) else: - preds[k] = preds[k].astype(paddle.float32) + preds[k] = paddle.to_tensor(preds[k], dtype='float32') elif isinstance(preds, list): for k in range(len(preds)): if isinstance(preds[k], dict): @@ -168,11 +169,12 @@ def to_float32(preds): elif isinstance(preds[k], list): preds[k] = to_float32(preds[k]) else: - preds[k] = preds[k].astype(paddle.float32) + preds[k] = paddle.to_tensor(preds[k], dtype='float32') else: - preds = preds.astype(paddle.float32) + preds[k] = paddle.to_tensor(preds[k], dtype='float32') return preds + def train(config, train_dataloader, valid_dataloader, From 1477b89aa40688210a0af9c6b819fd50ebdcc0fb Mon Sep 17 00:00:00 2001 From: andyjpaddle Date: Mon, 8 Aug 2022 08:59:30 +0000 Subject: [PATCH 24/25] fix amp train for re --- tools/infer/predict_det_eval.py | 363 ---------------------- tools/infer/predict_rec_eval.py | 534 -------------------------------- 2 files changed, 897 deletions(-) delete mode 100755 tools/infer/predict_det_eval.py delete mode 100755 tools/infer/predict_rec_eval.py diff --git a/tools/infer/predict_det_eval.py b/tools/infer/predict_det_eval.py deleted file mode 100755 index d1f832036b..0000000000 --- a/tools/infer/predict_det_eval.py +++ /dev/null @@ -1,363 +0,0 @@ -# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -import os -import sys - -__dir__ = os.path.dirname(os.path.abspath(__file__)) -sys.path.append(__dir__) -sys.path.insert(0, os.path.abspath(os.path.join(__dir__, '../..'))) - -os.environ["FLAGS_allocator_strategy"] = 'auto_growth' - -import cv2 -import numpy as np -import time -import sys - -import tools.infer.utility as utility -from ppocr.utils.logging import get_logger -from ppocr.utils.utility import get_image_file_list, check_and_read_gif -from ppocr.data import create_operators, transform -from ppocr.postprocess import build_post_process -import json -logger = get_logger() - - -class TextDetector(object): - def __init__(self, args): - self.args = args - self.det_algorithm = args.det_algorithm - self.use_onnx = args.use_onnx - pre_process_list = [{ - 'DetResizeForTest': { - 'limit_side_len': args.det_limit_side_len, - 'limit_type': args.det_limit_type, - } - }, { - 'NormalizeImage': { - 'std': [0.229, 0.224, 0.225], - 'mean': [0.485, 0.456, 0.406], - 'scale': '1./255.', - 'order': 'hwc' - } - }, { - 'ToCHWImage': None - }, { - 'KeepKeys': { - 'keep_keys': ['image', 'shape'] - } - }] - postprocess_params = {} - if self.det_algorithm == "DB": - postprocess_params['name'] = 'DBPostProcess' - postprocess_params["thresh"] = args.det_db_thresh - postprocess_params["box_thresh"] = args.det_db_box_thresh - postprocess_params["max_candidates"] = 1000 - postprocess_params["unclip_ratio"] = args.det_db_unclip_ratio - postprocess_params["use_dilation"] = args.use_dilation - postprocess_params["score_mode"] = args.det_db_score_mode - elif self.det_algorithm == "EAST": - postprocess_params['name'] = 'EASTPostProcess' - postprocess_params["score_thresh"] = args.det_east_score_thresh - postprocess_params["cover_thresh"] = args.det_east_cover_thresh - postprocess_params["nms_thresh"] = args.det_east_nms_thresh - elif self.det_algorithm == "SAST": - pre_process_list[0] = { - 'DetResizeForTest': { - 'resize_long': args.det_limit_side_len - } - } - postprocess_params['name'] = 'SASTPostProcess' - postprocess_params["score_thresh"] = args.det_sast_score_thresh - postprocess_params["nms_thresh"] = args.det_sast_nms_thresh - self.det_sast_polygon = args.det_sast_polygon - if self.det_sast_polygon: - postprocess_params["sample_pts_num"] = 6 - postprocess_params["expand_scale"] = 1.2 - postprocess_params["shrink_ratio_of_width"] = 0.2 - else: - postprocess_params["sample_pts_num"] = 2 - postprocess_params["expand_scale"] = 1.0 - postprocess_params["shrink_ratio_of_width"] = 0.3 - elif self.det_algorithm == "PSE": - postprocess_params['name'] = 'PSEPostProcess' - postprocess_params["thresh"] = args.det_pse_thresh - postprocess_params["box_thresh"] = args.det_pse_box_thresh - postprocess_params["min_area"] = args.det_pse_min_area - postprocess_params["box_type"] = args.det_pse_box_type - postprocess_params["scale"] = args.det_pse_scale - self.det_pse_box_type = args.det_pse_box_type - elif self.det_algorithm == "FCE": - pre_process_list[0] = { - 'DetResizeForTest': { - 'rescale_img': [1080, 736] - } - } - postprocess_params['name'] = 'FCEPostProcess' - postprocess_params["scales"] = args.scales - postprocess_params["alpha"] = args.alpha - postprocess_params["beta"] = args.beta - postprocess_params["fourier_degree"] = args.fourier_degree - postprocess_params["box_type"] = args.det_fce_box_type - else: - logger.info("unknown det_algorithm:{}".format(self.det_algorithm)) - sys.exit(0) - - self.preprocess_op = create_operators(pre_process_list) - self.postprocess_op = build_post_process(postprocess_params) - self.predictor, self.input_tensor, self.output_tensors, self.config = utility.create_predictor( - args, 'det', logger) - - if self.use_onnx: - img_h, img_w = self.input_tensor.shape[2:] - if img_h is not None and img_w is not None and img_h > 0 and img_w > 0: - pre_process_list[0] = { - 'DetResizeForTest': { - 'image_shape': [img_h, img_w] - } - } - self.preprocess_op = create_operators(pre_process_list) - - if args.benchmark: - import auto_log - pid = os.getpid() - gpu_id = utility.get_infer_gpuid() - self.autolog = auto_log.AutoLogger( - model_name="det", - model_precision=args.precision, - batch_size=1, - data_shape="dynamic", - save_path=None, - inference_config=self.config, - pids=pid, - process_name=None, - gpu_ids=gpu_id if args.use_gpu else None, - time_keys=[ - 'preprocess_time', 'inference_time', 'postprocess_time' - ], - warmup=2, - logger=logger) - - def order_points_clockwise(self, pts): - rect = np.zeros((4, 2), dtype="float32") - s = pts.sum(axis=1) - rect[0] = pts[np.argmin(s)] - rect[2] = pts[np.argmax(s)] - tmp = np.delete(pts, (np.argmin(s), np.argmax(s)), axis=0) - diff = np.diff(np.array(tmp), axis=1) - rect[1] = tmp[np.argmin(diff)] - rect[3] = tmp[np.argmax(diff)] - return rect - - def clip_det_res(self, points, img_height, img_width): - for pno in range(points.shape[0]): - points[pno, 0] = int(min(max(points[pno, 0], 0), img_width - 1)) - points[pno, 1] = int(min(max(points[pno, 1], 0), img_height - 1)) - return points - - def filter_tag_det_res(self, dt_boxes, image_shape): - img_height, img_width = image_shape[0:2] - dt_boxes_new = [] - for box in dt_boxes: - box = self.order_points_clockwise(box) - box = self.clip_det_res(box, img_height, img_width) - rect_width = int(np.linalg.norm(box[0] - box[1])) - rect_height = int(np.linalg.norm(box[0] - box[3])) - if rect_width <= 3 or rect_height <= 3: - continue - dt_boxes_new.append(box) - dt_boxes = np.array(dt_boxes_new) - return dt_boxes - - def filter_tag_det_res_only_clip(self, dt_boxes, image_shape): - img_height, img_width = image_shape[0:2] - dt_boxes_new = [] - for box in dt_boxes: - box = self.clip_det_res(box, img_height, img_width) - dt_boxes_new.append(box) - dt_boxes = np.array(dt_boxes_new) - return dt_boxes - - def __call__(self, img): - ori_im = img.copy() - data = {'image': img} - - st = time.time() - - if self.args.benchmark: - self.autolog.times.start() - - data = transform(data, self.preprocess_op) - img, shape_list = data - if img is None: - return None, 0 - img = np.expand_dims(img, axis=0) - shape_list = np.expand_dims(shape_list, axis=0) - img = img.copy() - - if self.args.benchmark: - self.autolog.times.stamp() - if self.use_onnx: - input_dict = {} - input_dict[self.input_tensor.name] = img - outputs = self.predictor.run(self.output_tensors, input_dict) - else: - self.input_tensor.copy_from_cpu(img) - self.predictor.run() - outputs = [] - for output_tensor in self.output_tensors: - output = output_tensor.copy_to_cpu() - outputs.append(output) - if self.args.benchmark: - self.autolog.times.stamp() - - preds = {} - if self.det_algorithm == "EAST": - preds['f_geo'] = outputs[0] - preds['f_score'] = outputs[1] - elif self.det_algorithm == 'SAST': - preds['f_border'] = outputs[0] - preds['f_score'] = outputs[1] - preds['f_tco'] = outputs[2] - preds['f_tvo'] = outputs[3] - elif self.det_algorithm in ['DB', 'PSE']: - preds['maps'] = outputs[0] - elif self.det_algorithm == 'FCE': - for i, output in enumerate(outputs): - preds['level_{}'.format(i)] = output - else: - raise NotImplementedError - - #self.predictor.try_shrink_memory() - post_result = self.postprocess_op(preds, shape_list) - dt_boxes = post_result[0]['points'] - if (self.det_algorithm == "SAST" and self.det_sast_polygon) or ( - self.det_algorithm in ["PSE", "FCE"] and - self.postprocess_op.box_type == 'poly'): - dt_boxes = self.filter_tag_det_res_only_clip(dt_boxes, ori_im.shape) - else: - dt_boxes = self.filter_tag_det_res(dt_boxes, ori_im.shape) - - if self.args.benchmark: - self.autolog.times.end(stamp=True) - et = time.time() - return dt_boxes, et - st - - -if __name__ == "__main__": - from ppocr.metrics.eval_det_iou import DetectionIoUEvaluator - evaluator = DetectionIoUEvaluator() - args = utility.parse_args() - - # image_file_list = get_image_file_list(args.image_dir) - def _check_image_file(path): - img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif'} - return any([path.lower().endswith(e) for e in img_end]) - - def get_image_file_list_from_txt(img_file): - imgs_lists = [] - label_lists = [] - if img_file is None or not os.path.exists(img_file): - raise Exception("not found any img file in {}".format(img_file)) - - img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif'} - root_dir = img_file.split('/')[0] - with open(img_file, 'r') as f: - lines = f.readlines() - for line in lines: - line = line.replace('\n', '').split('\t') - file_path, label = line[0], line[1] - file_path = os.path.join(root_dir, file_path) - if os.path.isfile(file_path) and _check_image_file(file_path): - imgs_lists.append(file_path) - label_lists.append(label) - - if len(imgs_lists) == 0: - raise Exception("not found any img file in {}".format(img_file)) - return imgs_lists, label_lists - - image_file_list, label_list = get_image_file_list_from_txt(args.image_dir) - - text_detector = TextDetector(args) - count = 0 - total_time = 0 - draw_img_save = "./inference_results" - - if args.warmup: - img = np.random.uniform(0, 255, [640, 640, 3]).astype(np.uint8) - for i in range(2): - res = text_detector(img) - - if not os.path.exists(draw_img_save): - os.makedirs(draw_img_save) - save_results = [] - results = [] - for idx in range(len(image_file_list)): - image_file = image_file_list[idx] - label = json.loads(label_list[idx]) - img, flag = check_and_read_gif(image_file) - if not flag: - img = cv2.imread(image_file) - if img is None: - logger.info("error in loading image:{}".format(image_file)) - continue - st = time.time() - dt_boxes, _ = text_detector(img) - elapse = time.time() - st - if count > 0: - total_time += elapse - count += 1 - save_pred = os.path.basename(image_file) + "\t" + str( - json.dumps([x.tolist() for x in dt_boxes])) + "\n" - save_results.append(save_pred) - - # for eval - gt_info_list = [] - det_info_list = [] - for dt_box in dt_boxes: - det_info = { - 'points': np.array( - dt_box, dtype=np.float32), - 'text': '' - } - det_info_list.append(det_info) - for lab in label: - gt_info = { - 'points': np.array( - lab['points'], dtype=np.float32), - 'text': '', - 'ignore': False - } - gt_info_list.append(gt_info) - result = evaluator.evaluate_image(gt_info_list, det_info_list) - results.append(result) - - metrics = evaluator.combine_results(results) - print('predict det eval on ', args.image_dir) - print('metrics: ', metrics) - -# logger.info(save_pred) -# logger.info("The predict time of {}: {}".format(image_file, elapse)) -# src_im = utility.draw_text_det_res(dt_boxes, image_file) -# img_name_pure = os.path.split(image_file)[-1] -# img_path = os.path.join(draw_img_save, -# "det_res_{}".format(img_name_pure)) -# cv2.imwrite(img_path, src_im) -# logger.info("The visualized image saved in {}".format(img_path)) - -# with open(os.path.join(draw_img_save, "det_results.txt"), 'w') as f: -# f.writelines(save_results) -# f.close() -# if args.benchmark: -# text_detector.autolog.report() diff --git a/tools/infer/predict_rec_eval.py b/tools/infer/predict_rec_eval.py deleted file mode 100755 index 3150d11ddf..0000000000 --- a/tools/infer/predict_rec_eval.py +++ /dev/null @@ -1,534 +0,0 @@ -# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -import os -import sys -from PIL import Image -__dir__ = os.path.dirname(os.path.abspath(__file__)) -sys.path.append(__dir__) -sys.path.insert(0, os.path.abspath(os.path.join(__dir__, '../..'))) - -os.environ["FLAGS_allocator_strategy"] = 'auto_growth' - -import cv2 -import numpy as np -import math -import time -import traceback -import paddle - -import tools.infer.utility as utility -from ppocr.postprocess import build_post_process -from ppocr.utils.logging import get_logger -from ppocr.utils.utility import get_image_file_list, check_and_read_gif - -logger = get_logger() - - -class TextRecognizer(object): - def __init__(self, args): - self.rec_image_shape = [int(v) for v in args.rec_image_shape.split(",")] - self.rec_batch_num = args.rec_batch_num - self.rec_algorithm = args.rec_algorithm - postprocess_params = { - 'name': 'CTCLabelDecode', - "character_dict_path": args.rec_char_dict_path, - "use_space_char": args.use_space_char - } - if self.rec_algorithm == "SRN": - postprocess_params = { - 'name': 'SRNLabelDecode', - "character_dict_path": args.rec_char_dict_path, - "use_space_char": args.use_space_char - } - elif self.rec_algorithm == "RARE": - postprocess_params = { - 'name': 'AttnLabelDecode', - "character_dict_path": args.rec_char_dict_path, - "use_space_char": args.use_space_char - } - elif self.rec_algorithm == 'NRTR': - postprocess_params = { - 'name': 'NRTRLabelDecode', - "character_dict_path": args.rec_char_dict_path, - "use_space_char": args.use_space_char - } - elif self.rec_algorithm == "SAR": - postprocess_params = { - 'name': 'SARLabelDecode', - "character_dict_path": args.rec_char_dict_path, - "use_space_char": args.use_space_char - } - elif self.rec_algorithm == 'ViTSTR': - postprocess_params = { - 'name': 'ViTSTRLabelDecode', - "character_dict_path": args.rec_char_dict_path, - "use_space_char": args.use_space_char - } - elif self.rec_algorithm == 'ABINet': - postprocess_params = { - 'name': 'ABINetLabelDecode', - "character_dict_path": args.rec_char_dict_path, - "use_space_char": args.use_space_char - } - self.postprocess_op = build_post_process(postprocess_params) - self.predictor, self.input_tensor, self.output_tensors, self.config = \ - utility.create_predictor(args, 'rec', logger) - self.benchmark = args.benchmark - self.use_onnx = args.use_onnx - if args.benchmark: - import auto_log - pid = os.getpid() - gpu_id = utility.get_infer_gpuid() - self.autolog = auto_log.AutoLogger( - model_name="rec", - model_precision=args.precision, - batch_size=args.rec_batch_num, - data_shape="dynamic", - save_path=None, #args.save_log_path, - inference_config=self.config, - pids=pid, - process_name=None, - gpu_ids=gpu_id if args.use_gpu else None, - time_keys=[ - 'preprocess_time', 'inference_time', 'postprocess_time' - ], - warmup=0, - logger=logger) - - def resize_norm_img(self, img, max_wh_ratio): - imgC, imgH, imgW = self.rec_image_shape - if self.rec_algorithm == 'NRTR' or self.rec_algorithm == 'ViTSTR': - img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) - # return padding_im - image_pil = Image.fromarray(np.uint8(img)) - if self.rec_algorithm == 'ViTSTR': - img = image_pil.resize([imgW, imgH], Image.BICUBIC) - else: - img = image_pil.resize([imgW, imgH], Image.ANTIALIAS) - img = np.array(img) - norm_img = np.expand_dims(img, -1) - norm_img = norm_img.transpose((2, 0, 1)) - if self.rec_algorithm == 'ViTSTR': - norm_img = norm_img.astype(np.float32) / 255. - else: - norm_img = norm_img.astype(np.float32) / 128. - 1. - return norm_img - - assert imgC == img.shape[2] - imgW = int((imgH * max_wh_ratio)) - if self.use_onnx: - w = self.input_tensor.shape[3:][0] - if w is not None and w > 0: - imgW = w - - h, w = img.shape[:2] - ratio = w / float(h) - if math.ceil(imgH * ratio) > imgW: - resized_w = imgW - else: - resized_w = int(math.ceil(imgH * ratio)) - if self.rec_algorithm == 'RARE': - if resized_w > self.rec_image_shape[2]: - resized_w = self.rec_image_shape[2] - imgW = self.rec_image_shape[2] - resized_image = cv2.resize(img, (resized_w, imgH)) - resized_image = resized_image.astype('float32') - resized_image = resized_image.transpose((2, 0, 1)) / 255 - resized_image -= 0.5 - resized_image /= 0.5 - padding_im = np.zeros((imgC, imgH, imgW), dtype=np.float32) - padding_im[:, :, 0:resized_w] = resized_image - return padding_im - - def resize_norm_img_srn(self, img, image_shape): - imgC, imgH, imgW = image_shape - - img_black = np.zeros((imgH, imgW)) - im_hei = img.shape[0] - im_wid = img.shape[1] - - if im_wid <= im_hei * 1: - img_new = cv2.resize(img, (imgH * 1, imgH)) - elif im_wid <= im_hei * 2: - img_new = cv2.resize(img, (imgH * 2, imgH)) - elif im_wid <= im_hei * 3: - img_new = cv2.resize(img, (imgH * 3, imgH)) - else: - img_new = cv2.resize(img, (imgW, imgH)) - - img_np = np.asarray(img_new) - img_np = cv2.cvtColor(img_np, cv2.COLOR_BGR2GRAY) - img_black[:, 0:img_np.shape[1]] = img_np - img_black = img_black[:, :, np.newaxis] - - row, col, c = img_black.shape - c = 1 - - return np.reshape(img_black, (c, row, col)).astype(np.float32) - - def srn_other_inputs(self, image_shape, num_heads, max_text_length): - - imgC, imgH, imgW = image_shape - feature_dim = int((imgH / 8) * (imgW / 8)) - - encoder_word_pos = np.array(range(0, feature_dim)).reshape( - (feature_dim, 1)).astype('int64') - gsrm_word_pos = np.array(range(0, max_text_length)).reshape( - (max_text_length, 1)).astype('int64') - - gsrm_attn_bias_data = np.ones((1, max_text_length, max_text_length)) - gsrm_slf_attn_bias1 = np.triu(gsrm_attn_bias_data, 1).reshape( - [-1, 1, max_text_length, max_text_length]) - gsrm_slf_attn_bias1 = np.tile( - gsrm_slf_attn_bias1, - [1, num_heads, 1, 1]).astype('float32') * [-1e9] - - gsrm_slf_attn_bias2 = np.tril(gsrm_attn_bias_data, -1).reshape( - [-1, 1, max_text_length, max_text_length]) - gsrm_slf_attn_bias2 = np.tile( - gsrm_slf_attn_bias2, - [1, num_heads, 1, 1]).astype('float32') * [-1e9] - - encoder_word_pos = encoder_word_pos[np.newaxis, :] - gsrm_word_pos = gsrm_word_pos[np.newaxis, :] - - return [ - encoder_word_pos, gsrm_word_pos, gsrm_slf_attn_bias1, - gsrm_slf_attn_bias2 - ] - - def process_image_srn(self, img, image_shape, num_heads, max_text_length): - norm_img = self.resize_norm_img_srn(img, image_shape) - norm_img = norm_img[np.newaxis, :] - - [encoder_word_pos, gsrm_word_pos, gsrm_slf_attn_bias1, gsrm_slf_attn_bias2] = \ - self.srn_other_inputs(image_shape, num_heads, max_text_length) - - gsrm_slf_attn_bias1 = gsrm_slf_attn_bias1.astype(np.float32) - gsrm_slf_attn_bias2 = gsrm_slf_attn_bias2.astype(np.float32) - encoder_word_pos = encoder_word_pos.astype(np.int64) - gsrm_word_pos = gsrm_word_pos.astype(np.int64) - - return (norm_img, encoder_word_pos, gsrm_word_pos, gsrm_slf_attn_bias1, - gsrm_slf_attn_bias2) - - def resize_norm_img_sar(self, img, image_shape, - width_downsample_ratio=0.25): - imgC, imgH, imgW_min, imgW_max = image_shape - h = img.shape[0] - w = img.shape[1] - valid_ratio = 1.0 - # make sure new_width is an integral multiple of width_divisor. - width_divisor = int(1 / width_downsample_ratio) - # resize - ratio = w / float(h) - resize_w = math.ceil(imgH * ratio) - if resize_w % width_divisor != 0: - resize_w = round(resize_w / width_divisor) * width_divisor - if imgW_min is not None: - resize_w = max(imgW_min, resize_w) - if imgW_max is not None: - valid_ratio = min(1.0, 1.0 * resize_w / imgW_max) - resize_w = min(imgW_max, resize_w) - resized_image = cv2.resize(img, (resize_w, imgH)) - resized_image = resized_image.astype('float32') - # norm - if image_shape[0] == 1: - resized_image = resized_image / 255 - resized_image = resized_image[np.newaxis, :] - else: - resized_image = resized_image.transpose((2, 0, 1)) / 255 - resized_image -= 0.5 - resized_image /= 0.5 - resize_shape = resized_image.shape - padding_im = -1.0 * np.ones((imgC, imgH, imgW_max), dtype=np.float32) - padding_im[:, :, 0:resize_w] = resized_image - pad_shape = padding_im.shape - - return padding_im, resize_shape, pad_shape, valid_ratio - - def resize_norm_img_svtr(self, img, image_shape): - - imgC, imgH, imgW = image_shape - resized_image = cv2.resize( - img, (imgW, imgH), interpolation=cv2.INTER_LINEAR) - resized_image = resized_image.astype('float32') - resized_image = resized_image.transpose((2, 0, 1)) / 255 - resized_image -= 0.5 - resized_image /= 0.5 - return resized_image - - def resize_norm_img_abinet(self, img, image_shape): - - imgC, imgH, imgW = image_shape - - resized_image = cv2.resize( - img, (imgW, imgH), interpolation=cv2.INTER_LINEAR) - resized_image = resized_image.astype('float32') - resized_image = resized_image / 255. - - mean = np.array([0.485, 0.456, 0.406]) - std = np.array([0.229, 0.224, 0.225]) - resized_image = ( - resized_image - mean[None, None, ...]) / std[None, None, ...] - resized_image = resized_image.transpose((2, 0, 1)) - resized_image = resized_image.astype('float32') - - return resized_image - - def __call__(self, img_list): - img_num = len(img_list) - # Calculate the aspect ratio of all text bars - width_list = [] - for img in img_list: - width_list.append(img.shape[1] / float(img.shape[0])) - # Sorting can speed up the recognition process - indices = np.argsort(np.array(width_list)) - rec_res = [['', 0.0]] * img_num - batch_num = self.rec_batch_num - st = time.time() - if self.benchmark: - self.autolog.times.start() - for beg_img_no in range(0, img_num, batch_num): - end_img_no = min(img_num, beg_img_no + batch_num) - norm_img_batch = [] - imgC, imgH, imgW = self.rec_image_shape - max_wh_ratio = imgW / imgH - # max_wh_ratio = 0 - for ino in range(beg_img_no, end_img_no): - h, w = img_list[indices[ino]].shape[0:2] - wh_ratio = w * 1.0 / h - max_wh_ratio = max(max_wh_ratio, wh_ratio) - for ino in range(beg_img_no, end_img_no): - - if self.rec_algorithm == "SAR": - norm_img, _, _, valid_ratio = self.resize_norm_img_sar( - img_list[indices[ino]], self.rec_image_shape) - norm_img = norm_img[np.newaxis, :] - valid_ratio = np.expand_dims(valid_ratio, axis=0) - valid_ratios = [] - valid_ratios.append(valid_ratio) - norm_img_batch.append(norm_img) - elif self.rec_algorithm == "SRN": - norm_img = self.process_image_srn( - img_list[indices[ino]], self.rec_image_shape, 8, 25) - encoder_word_pos_list = [] - gsrm_word_pos_list = [] - gsrm_slf_attn_bias1_list = [] - gsrm_slf_attn_bias2_list = [] - encoder_word_pos_list.append(norm_img[1]) - gsrm_word_pos_list.append(norm_img[2]) - gsrm_slf_attn_bias1_list.append(norm_img[3]) - gsrm_slf_attn_bias2_list.append(norm_img[4]) - norm_img_batch.append(norm_img[0]) - elif self.rec_algorithm == "SVTR": - norm_img = self.resize_norm_img_svtr(img_list[indices[ino]], - self.rec_image_shape) - norm_img = norm_img[np.newaxis, :] - norm_img_batch.append(norm_img) - elif self.rec_algorithm == "ABINet": - norm_img = self.resize_norm_img_abinet( - img_list[indices[ino]], self.rec_image_shape) - norm_img = norm_img[np.newaxis, :] - norm_img_batch.append(norm_img) - else: - norm_img = self.resize_norm_img(img_list[indices[ino]], - max_wh_ratio) - norm_img = norm_img[np.newaxis, :] - norm_img_batch.append(norm_img) - norm_img_batch = np.concatenate(norm_img_batch) - norm_img_batch = norm_img_batch.copy() - if self.benchmark: - self.autolog.times.stamp() - - if self.rec_algorithm == "SRN": - encoder_word_pos_list = np.concatenate(encoder_word_pos_list) - gsrm_word_pos_list = np.concatenate(gsrm_word_pos_list) - gsrm_slf_attn_bias1_list = np.concatenate( - gsrm_slf_attn_bias1_list) - gsrm_slf_attn_bias2_list = np.concatenate( - gsrm_slf_attn_bias2_list) - - inputs = [ - norm_img_batch, - encoder_word_pos_list, - gsrm_word_pos_list, - gsrm_slf_attn_bias1_list, - gsrm_slf_attn_bias2_list, - ] - if self.use_onnx: - input_dict = {} - input_dict[self.input_tensor.name] = norm_img_batch - outputs = self.predictor.run(self.output_tensors, - input_dict) - preds = {"predict": outputs[2]} - else: - input_names = self.predictor.get_input_names() - for i in range(len(input_names)): - input_tensor = self.predictor.get_input_handle( - input_names[i]) - input_tensor.copy_from_cpu(inputs[i]) - self.predictor.run() - outputs = [] - for output_tensor in self.output_tensors: - output = output_tensor.copy_to_cpu() - outputs.append(output) - if self.benchmark: - self.autolog.times.stamp() - preds = {"predict": outputs[2]} - elif self.rec_algorithm == "SAR": - valid_ratios = np.concatenate(valid_ratios) - inputs = [ - norm_img_batch, - valid_ratios, - ] - if self.use_onnx: - input_dict = {} - input_dict[self.input_tensor.name] = norm_img_batch - outputs = self.predictor.run(self.output_tensors, - input_dict) - preds = outputs[0] - else: - input_names = self.predictor.get_input_names() - for i in range(len(input_names)): - input_tensor = self.predictor.get_input_handle( - input_names[i]) - input_tensor.copy_from_cpu(inputs[i]) - self.predictor.run() - outputs = [] - for output_tensor in self.output_tensors: - output = output_tensor.copy_to_cpu() - outputs.append(output) - if self.benchmark: - self.autolog.times.stamp() - preds = outputs[0] - else: - if self.use_onnx: - input_dict = {} - input_dict[self.input_tensor.name] = norm_img_batch - outputs = self.predictor.run(self.output_tensors, - input_dict) - preds = outputs[0] - else: - self.input_tensor.copy_from_cpu(norm_img_batch) - self.predictor.run() - outputs = [] - for output_tensor in self.output_tensors: - output = output_tensor.copy_to_cpu() - outputs.append(output) - if self.benchmark: - self.autolog.times.stamp() - if len(outputs) != 1: - preds = outputs - else: - preds = outputs[0] - rec_result = self.postprocess_op(preds) - for rno in range(len(rec_result)): - rec_res[indices[beg_img_no + rno]] = rec_result[rno] - if self.benchmark: - self.autolog.times.end(stamp=True) - return rec_res, time.time() - st - - -def main(args): - # image_file_list = get_image_file_list(args.image_dir) - - def _check_image_file(path): - img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif'} - return any([path.lower().endswith(e) for e in img_end]) - - def get_image_file_list_from_txt(img_file): - imgs_lists = [] - label_lists = [] - if img_file is None or not os.path.exists(img_file): - raise Exception("not found any img file in {}".format(img_file)) - - img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif'} - root_dir = img_file.split('/')[0] - with open(img_file, 'r') as f: - lines = f.readlines() - for line in lines: - line = line.replace('\n', '').split('\t') - file_path, label = line[0], line[1] - file_path = os.path.join(root_dir, file_path) - if os.path.isfile(file_path) and _check_image_file(file_path): - imgs_lists.append(file_path) - label_lists.append(label) - - if len(imgs_lists) == 0: - raise Exception("not found any img file in {}".format(img_file)) - return imgs_lists, label_lists - - image_file_list, label_list = get_image_file_list_from_txt(args.image_dir) - - text_recognizer = TextRecognizer(args) - valid_image_file_list = [] - img_list = [] - - logger.info( - "In PP-OCRv3, rec_image_shape parameter defaults to '3, 48, 320', " - "if you are using recognition model with PP-OCRv2 or an older version, please set --rec_image_shape='3,32,320" - ) - # warmup 2 times - if args.warmup: - img = np.random.uniform(0, 255, [48, 320, 3]).astype(np.uint8) - for i in range(2): - res = text_recognizer([img] * int(args.rec_batch_num)) - - for image_file in image_file_list: - img, flag = check_and_read_gif(image_file) - if not flag: - img = cv2.imread(image_file) - if img is None: - logger.info("error in loading image:{}".format(image_file)) - continue - valid_image_file_list.append(image_file) - img_list.append(img) - - try: - rec_res, _ = text_recognizer(img_list) - except Exception as E: - logger.info(traceback.format_exc()) - logger.info(E) - exit() - correct_num = 0 - for ino in range(len(img_list)): - pred = rec_res[ino][0] - gt = label_list[ino] - if pred == gt: - correct_num += 1 - acc = correct_num * 1.0 / len(img_list) - print('predict rec eval on ', args.image_dir) - print('acc: ', acc) - - # for debug bad case - bad_case_lines = [] - for ino in range(len(img_list)): - pred = rec_res[ino][0] - gt = label_list[ino] - if pred != gt and len(gt) <= 25: - bad_case = valid_image_file_list[ - ino] + '\t' + 'pred:' + pred + '\t' + 'gt:' + gt + '\n' - bad_case_lines.append(bad_case) - - with open('bad_case_hwdb2.txt', 'a+') as f: - f.writelines(bad_case_lines) - # end debug case - - if args.benchmark: - text_recognizer.autolog.report() - - -if __name__ == "__main__": - main(utility.parse_args()) From 8e3aa3a09b60c3ca99ad20dc3ddb8150efd28f65 Mon Sep 17 00:00:00 2001 From: andyjpaddle Date: Mon, 8 Aug 2022 09:02:11 +0000 Subject: [PATCH 25/25] fix amp train for re --- tools/program.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/program.py b/tools/program.py index dd8037e473..53994b8560 100755 --- a/tools/program.py +++ b/tools/program.py @@ -171,7 +171,7 @@ def to_float32(preds): else: preds[k] = paddle.to_tensor(preds[k], dtype='float32') else: - preds[k] = paddle.to_tensor(preds[k], dtype='float32') + preds = paddle.to_tensor(preds, dtype='float32') return preds