Merge pull request #7879 from WenmuZhou/tipc_2

[TIPC] add kl and pact for slanet and vi-layoutxlm

ppstructure/docs/quickstart.md

@@ -214,7 +214,7 @@ for line in result:
 <a name="225"></a>
 #### 2.2.5 关键信息抽取
 
-关键信息抽取暂不支持通过whl包调用，详细使用教程请参考：[关键信息抽取教程](../kie/README_ch.md)。
+关键信息抽取暂不支持通过whl包调用，详细使用教程请参考：[inference文档](./inference.md)。
 
 <a name="226"></a>
 

ppstructure/docs/quickstart_en.md

@@ -94,7 +94,7 @@ paddleocr --image_dir=ppstructure/docs/table/table.jpg --type=structure --layout
 
 #### 2.1.5 Key Information Extraction
 
-Key information extraction does not currently support use by the whl package. For detailed usage tutorials, please refer to: [Key Information Extraction](../kie/README.md).
+Key information extraction does not currently support use by the whl package. For detailed usage tutorials, please refer to: [inference document](./inference_en.md).
 
 <a name="216"></a>
 #### 2.1.6 layout recovery

test_tipc/configs/slanet/SLANet.yml

@@ -12,7 +12,7 @@ Global:
   checkpoints:
   save_inference_dir: ./output/SLANet/infer
   use_visualdl: False
-  infer_img: doc/table/table.jpg
+  infer_img: ppstructure/docs/table/table.jpg
   # for data or label process
   character_dict_path: ppocr/utils/dict/table_structure_dict.txt
   character_type: en

test_tipc/configs/slanet/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt

+===========================train_params===========================
+model_name:slanet
+python:python3.7
+gpu_list:0|0,1
+Global.use_gpu:True|True
+Global.auto_cast:amp
+Global.epoch_num:lite_train_lite_infer=3|whole_train_whole_infer=50
+Global.save_model_dir:./output/
+Train.loader.batch_size_per_card:lite_train_lite_infer=16|whole_train_whole_infer=128
+Global.pretrained_model:./pretrain_models/en_ppstructure_mobile_v2.0_SLANet_train/best_accuracy
+train_model_name:latest
+train_infer_img_dir:./ppstructure/docs/table/table.jpg
+null:null
+##
+trainer:norm_train
+norm_train:tools/train.py -c test_tipc/configs/slanet/SLANet.yml -o
+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/
+Global.checkpoints:
+norm_export:tools/export_model.py -c test_tipc/configs/slanet/SLANet.yml -o
+quant_export:
+fpgm_export: 
+distill_export:null
+export1:null
+export2:null
+##
+infer_model:./inference/en_ppstructure_mobile_v2.0_SLANet_train
+infer_export:null
+infer_quant:False
+inference:ppstructure/table/predict_table.py --det_model_dir=./inference/en_ppocr_mobile_v2.0_table_det_infer --rec_model_dir=./inference/en_ppocr_mobile_v2.0_table_rec_infer  --rec_char_dict_path=./ppocr/utils/dict/table_dict.txt --table_char_dict_path=./ppocr/utils/dict/table_structure_dict.txt --image_dir=./ppstructure/docs/table/table.jpg --det_limit_side_len=736 --det_limit_type=min --output ./output/table
+--use_gpu:True|False
+--enable_mkldnn:False
+--cpu_threads:6
+--rec_batch_num:1
+--use_tensorrt:True
+--precision:fp32
+--table_model_dir:
+--image_dir:./ppstructure/docs/table/table.jpg
+null:null
+--benchmark:True
+null:null
+===========================infer_benchmark_params==========================
+random_infer_input:[{float32,[3,488,488]}]

test_tipc/configs/slanet/train_pact_infer_python.txt

+===========================train_params===========================
+model_name:slanet_PACT
+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=50
+Global.save_model_dir:./output/
+Train.loader.batch_size_per_card:lite_train_lite_infer=2|whole_train_whole_infer=2
+Global.pretrained_model:./pretrain_models/en_ppstructure_mobile_v2.0_SLANet_train/best_accuracy
+train_model_name:latest
+train_infer_img_dir:./ppstructure/docs/table/table.jpg
+null:null
+##
+trainer:pact_train
+norm_train:null 
+pact_train:deploy/slim/quantization/quant.py -c test_tipc/configs/slanet/SLANet.yml  -o 
+fpgm_train:null
+distill_train:null
+null:null
+null:null
+##
+===========================eval_params=========================== 
+eval:null
+null:null
+##
+===========================infer_params===========================
+Global.save_inference_dir:./output/
+Global.checkpoints:
+norm_export:null 
+quant_export:deploy/slim/quantization/export_model.py -c test_tipc/configs/slanet/SLANet.yml  -o 
+fpgm_export: 
+distill_export:null
+export1:null
+export2:null
+##
+infer_model:./inference/en_ppocr_mobile_v2.0_table_structure_infer
+infer_export:null
+infer_quant:True
+inference:ppstructure/table/predict_table.py --det_model_dir=./inference/en_ppocr_mobile_v2.0_table_det_infer --rec_model_dir=./inference/en_ppocr_mobile_v2.0_table_rec_infer  --rec_char_dict_path=./ppocr/utils/dict/table_dict.txt --table_char_dict_path=./ppocr/utils/dict/table_structure_dict.txt --image_dir=./ppstructure/docs/table/table.jpg --det_limit_side_len=736 --det_limit_type=min --output ./output/table
+--use_gpu:True|False
+--enable_mkldnn:False
+--cpu_threads:6
+--rec_batch_num:1
+--use_tensorrt:False
+--precision:fp32
+--table_model_dir:
+--image_dir:./ppstructure/docs/table/table.jpg
+null:null
+--benchmark:True
+null:null
+===========================infer_benchmark_params==========================
+random_infer_input:[{float32,[3,488,488]}]

test_tipc/configs/slanet/train_ptq_infer_python.txt

+===========================train_params===========================
+model_name:slanet_KL
+python:python3.7
+Global.pretrained_model:
+Global.save_inference_dir:null
+infer_model:./inference/en_ppstructure_mobile_v2.0_SLANet_infer/
+infer_export:deploy/slim/quantization/quant_kl.py -c test_tipc/configs/slanet/SLANet.yml -o
+infer_quant:True
+inference:ppstructure/table/predict_table.py --det_model_dir=./inference/ch_PP-OCRv3_det_infer --rec_model_dir=./inference/ch_PP-OCRv3_rec_infer  --rec_char_dict_path=./ppocr/utils/ppocr_keys_v1.txt --table_char_dict_path=./ppocr/utils/dict/table_structure_dict.txt --image_dir=./ppstructure/docs/table/table.jpg --det_limit_side_len=736 --det_limit_type=min --output ./output/table
+--use_gpu:True|False
+--enable_mkldnn:False
+--cpu_threads:6
+--rec_batch_num:1
+--use_tensorrt:False
+--precision:int8
+--table_model_dir:
+--image_dir:./ppstructure/docs/table/table.jpg
+null:null
+--benchmark:True
+null:null
+null:null

test_tipc/configs/vi_layoutxlm_ser/train_linux_gpu_fleet_normal_infer_python_linux_gpu_cpu.txt

+===========================train_params===========================
+model_name:vi_layoutxlm_ser
+python:python3.7
+gpu_list:192.168.0.1,192.168.0.2;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=4|whole_train_whole_infer=8
+Architecture.Backbone.checkpoints:null
+train_model_name:latest
+train_infer_img_dir:ppstructure/docs/kie/input/zh_val_42.jpg
+null:null
+##
+trainer:norm_train
+norm_train:tools/train.py -c ./configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh.yml -o 
+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/kie/vi_layoutxlm/ser_vi_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/kie/predict_kie_token_ser.py  --kie_algorithm=LayoutXLM  --ser_dict_path=train_data/XFUND/class_list_xfun.txt --output=output --ocr_order_method=tb-yx
+--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/kie/input/zh_val_42.jpg
+null:null
+--benchmark:True
+null:null
+===========================infer_benchmark_params==========================
+random_infer_input:[{float32,[3,224,224]}]

test_tipc/configs/vi_layoutxlm_ser/train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt

+===========================train_params===========================
+model_name:vi_layoutxlm_ser
+python:python3.7
+gpu_list:0|0,1
+Global.use_gpu:True|True
+Global.auto_cast:amp
+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=4|whole_train_whole_infer=8
+Architecture.Backbone.checkpoints:null
+train_model_name:latest
+train_infer_img_dir:ppstructure/docs/kie/input/zh_val_42.jpg
+null:null
+##
+trainer:norm_train
+norm_train:tools/train.py -c ./configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh.yml -o
+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/kie/vi_layoutxlm/ser_vi_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/kie/predict_kie_token_ser.py  --kie_algorithm=LayoutXLM  --ser_dict_path=train_data/XFUND/class_list_xfun.txt --output=output --ocr_order_method=tb-yx
+--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/kie/input/zh_val_42.jpg
+null:null
+--benchmark:True
+null:null
+===========================infer_benchmark_params==========================
+random_infer_input:[{float32,[3,224,224]}]

test_tipc/configs/vi_layoutxlm_ser/train_pact_infer_python.txt

+===========================train_params===========================
+model_name:vi_layoutxlm_ser_PACT
+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=4|whole_train_whole_infer=8
+Architecture.Backbone.pretrained:./pretrain_models/ser_vi_layoutxlm_xfund_pretrained/best_accuracy
+train_model_name:latest
+train_infer_img_dir:ppstructure/docs/kie/input/zh_val_42.jpg
+null:null
+##
+trainer:pact_train
+norm_train:null
+pact_train:deploy/slim/quantization/quant.py -c ./configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh.yml -o Global.eval_batch_step=[2000,10]
+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:null
+quant_export:deploy/slim/quantization/export_model.py -c ./configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh.yml -o 
+fpgm_export: null
+distill_export:null
+export1:null
+export2:null
+##
+infer_model:null
+infer_export:null
+infer_quant:False
+inference:ppstructure/kie/predict_kie_token_ser.py  --kie_algorithm=LayoutXLM  --ser_dict_path=train_data/XFUND/class_list_xfun.txt --output=output --ocr_order_method=tb-yx
+--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/kie/input/zh_val_42.jpg
+null:null
+--benchmark:True
+null:null
+===========================infer_benchmark_params==========================
+random_infer_input:[{float32,[3,224,224]}]

test_tipc/configs/vi_layoutxlm_ser/train_ptq_infer_python.txt

+===========================train_params===========================
+model_name:vi_layoutxlm_ser_KL
+python:python3.7
+Global.pretrained_model:
+Global.save_inference_dir:null
+infer_model:./inference/ser_vi_layoutxlm_xfund_infer/
+infer_export:deploy/slim/quantization/quant_kl.py -c ./configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh.yml -o Train.loader.batch_size_per_card=1 Eval.loader.batch_size_per_card=1
+infer_quant:True
+inference:ppstructure/kie/predict_kie_token_ser.py  --kie_algorithm=LayoutXLM  --ser_dict_path=train_data/XFUND/class_list_xfun.txt --output=output --ocr_order_method=tb-yx
+--use_gpu:True|False
+--enable_mkldnn:False
+--cpu_threads:6
+--rec_batch_num:1
+--use_tensorrt:False
+--precision:int8
+--ser_model_dir:
+--image_dir:./ppstructure/docs/kie/input/zh_val_42.jpg
+null:null
+--benchmark:True
+null:null
+null:null

test_tipc/prepare.sh

@@ -164,7 +164,7 @@ if [ ${MODE} = "lite_train_lite_infer" ];then
         wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_rec_infer.tar --no-check-certificate
         cd ./inference/ && tar xf en_ppocr_mobile_v2.0_table_det_infer.tar && tar xf en_ppocr_mobile_v2.0_table_rec_infer.tar && cd ../
     fi
-    if [ ${model_name} == "slanet" ];then
+    if [[ ${model_name} =~ "slanet" ]];then
         wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/ppstructure/models/slanet/en_ppstructure_mobile_v2.0_SLANet_train.tar --no-check-certificate
         cd ./pretrain_models/ && tar xf en_ppstructure_mobile_v2.0_SLANet_train.tar  && cd ../
         wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_det_infer.tar --no-check-certificate
@@ -267,12 +267,16 @@ if [ ${MODE} = "lite_train_lite_infer" ];then
         wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/pplayout/ser_LayoutXLM_xfun_zh.tar --no-check-certificate
         cd ./pretrain_models/ && tar xf ser_LayoutXLM_xfun_zh.tar  && cd ../
     fi
-    if [ ${model_name} == "vi_layoutxlm_ser" ]; then
+    if [[ ${model_name} =~ "vi_layoutxlm_ser" ]]; then
         ${python_name} -m pip install -r ppstructure/kie/requirements.txt
         ${python_name} -m pip install opencv-python -U
         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 ../
+        if [ ${model_name} == "vi_layoutxlm_ser_PACT" ]; then
+            wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/ppstructure/models/vi_layoutxlm/ser_vi_layoutxlm_xfund_pretrained.tar --no-check-certificate
+            cd ./pretrain_models/ && tar xf ser_vi_layoutxlm_xfund_pretrained.tar  && cd ../
+        fi
     fi
     if [ ${model_name} == "det_r18_ct" ]; then
         wget -nc -P ./pretrain_models/  https://paddleocr.bj.bcebos.com/pretrained/ResNet18_vd_pretrained.pdparams  --no-check-certificate
@@ -532,6 +536,18 @@ elif [ ${MODE} = "whole_infer" ];then
         fi
         cd ../
     fi
+    if [[ ${model_name} =~ "slanet" ]];then
+        wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/ppstructure/models/slanet/en_ppstructure_mobile_v2.0_SLANet_infer.tar --no-check-certificate
+        wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_infer.tar --no-check-certificate
+        wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_infer.tar --no-check-certificate
+        cd ./inference/ && tar xf en_ppstructure_mobile_v2.0_SLANet_infer.tar && tar xf ch_PP-OCRv3_det_infer.tar && tar xf ch_PP-OCRv3_rec_infer.tar && cd ../
+    fi
+    if [[ ${model_name} =~ "vi_layoutxlm_ser" ]]; then
+        ${python_name} -m pip install -r ppstructure/kie/requirements.txt
+        ${python_name} -m pip install opencv-python -U 
+        wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/ppstructure/models/vi_layoutxlm/ser_vi_layoutxlm_xfund_infer.tar --no-check-certificate
+        cd ./inference/ && tar xf ser_vi_layoutxlm_xfund_infer.tar & cd ../
+    fi
     if [[ ${model_name} =~ "layoutxlm_ser" ]]; then
         ${python_name} -m pip install -r ppstructure/kie/requirements.txt
         ${python_name} -m pip install opencv-python -U