Unverified Commit 3d695fcc authored by xiaoting's avatar xiaoting Committed by GitHub
Browse files

Merge branch 'dygraph' into add_test_serving

parents ae12e0d5 d2d1483c
# 从训练到推理部署工具链测试方法介绍
# 推理部署导航
test.sh和params.txt文件配合使用,完成OCR轻量检测和识别模型从训练到预测的流程测试。
飞桨除了基本的模型训练和预测,还提供了支持多端多平台的高性能推理部署工具。本文档提供了PaddleOCR中所有模型的推理部署导航,方便用户查阅每种模型的推理部署打通情况,并可以进行一键测试。
# 安装依赖
- 安装PaddlePaddle >= 2.0
- 安装PaddleOCR依赖
```
pip3 install -r ../requirements.txt
```
- 安装autolog
```
git clone https://github.com/LDOUBLEV/AutoLog
cd AutoLog
pip3 install -r requirements.txt
python3 setup.py bdist_wheel
pip3 install ./dist/auto_log-1.0.0-py3-none-any.whl
cd ../
```
<div align="center">
<img src="docs/guide.png" width="1000">
</div>
# 目录介绍
打通情况汇总如下,已填写的部分表示可以使用本工具进行一键测试,未填写的表示正在支持中。
```bash
tests/
├── ocr_det_params.txt # 测试OCR检测模型的参数配置文件
├── ocr_rec_params.txt # 测试OCR识别模型的参数配置文件
├── ocr_ppocr_mobile_params.txt # 测试OCR检测+识别模型串联的参数配置文件
└── prepare.sh # 完成test.sh运行所需要的数据和模型下载
└── test.sh # 测试主程序
```
| 算法论文 | 模型名称 | 模型类型 | python训练预测 | 其他 |
| :--- | :--- | :---- | :-------- | :---- |
| DB |ch_ppocr_mobile_v2.0_det | 检测 | 支持 | Paddle Inference: C++预测 <br> Paddle Serving: Python, C++ <br> Paddle-Lite: Python, C++ / ARM CPU |
| DB |ch_ppocr_server_v2.0_det | 检测 | 支持 | Paddle Inference: C++预测 <br> Paddle Serving: Python, C++ <br> Paddle-Lite: Python, C++ / ARM CPU |
| DB |ch_PP-OCRv2_det | 检测 |
| CRNN |ch_ppocr_mobile_v2.0_rec | 识别 | 支持 | Paddle Inference: C++预测 <br> Paddle Serving: Python, C++ <br> Paddle-Lite: Python, C++ / ARM CPU |
| CRNN |ch_ppocr_server_v2.0_rec | 识别 | 支持 | Paddle Inference: C++预测 <br> Paddle Serving: Python, C++ <br> Paddle-Lite: Python, C++ / ARM CPU |
| CRNN |ch_PP-OCRv2_rec | 识别 |
| DB |det_mv3_db_v2.0 | 检测 |
| DB |det_r50_vd_db_v2.0 | 检测 |
| EAST |det_mv3_east_v2.0 | 检测 |
| EAST |det_r50_vd_east_v2.0 | 检测 |
| PSENet |det_mv3_pse_v2.0 | 检测 |
| PSENet |det_r50_vd_pse_v2.0 | 检测 |
| SAST |det_r50_vd_sast_totaltext_v2.0 | 检测 |
| Rosetta|rec_mv3_none_none_ctc_v2.0 | 识别 |
| Rosetta|rec_r34_vd_none_none_ctc_v2.0 | 识别 |
| CRNN |rec_mv3_none_bilstm_ctc_v2.0 | 识别 |
| CRNN |rec_r34_vd_none_bilstm_ctc_v2.0| 识别 |
| StarNet|rec_mv3_tps_bilstm_ctc_v2.0 | 识别 |
| StarNet|rec_r34_vd_tps_bilstm_ctc_v2.0 | 识别 |
| RARE |rec_mv3_tps_bilstm_att_v2.0 | 识别 |
| RARE |rec_r34_vd_tps_bilstm_att_v2.0 | 识别 |
| SRN |rec_r50fpn_vd_none_srn | 识别 |
| NRTR |rec_mtb_nrtr | 识别 |
| SAR |rec_r31_sar | 识别 |
| PGNet |rec_r34_vd_none_none_ctc_v2.0 | 端到端|
# 使用方法
test.sh包含四种运行模式,每种模式的运行数据不同,分别用于测试速度和精度,分别是:
- 模式1:lite_train_infer,使用少量数据训练,用于快速验证训练到预测的走通流程,不验证精度和速度;
```shell
bash tests/prepare.sh ./tests/ocr_det_params.txt 'lite_train_infer'
bash tests/test.sh ./tests/ocr_det_params.txt 'lite_train_infer'
```
- 模式2:whole_infer,使用少量数据训练,一定量数据预测,用于验证训练后的模型执行预测,预测速度是否合理;
```shell
bash tests/prepare.sh ./tests/ocr_det_params.txt 'whole_infer'
bash tests/test.sh ./tests/ocr_det_params.txt 'whole_infer'
```
## 一键测试工具使用
### 目录介绍
- 模式3:infer 不训练,全量数据预测,走通开源模型评估、动转静,检查inference model预测时间和精度;
```shell
bash tests/prepare.sh ./tests/ocr_det_params.txt 'infer'
# 用法1:
bash tests/test.sh ./tests/ocr_det_params.txt 'infer'
# 用法2: 指定GPU卡预测,第三个传入参数为GPU卡号
bash tests/test.sh ./tests/ocr_det_params.txt 'infer' '1'
```
tests/
├── configs/ # 配置文件目录
├── det_mv3_db.yml # 测试mobile版ppocr检测模型训练的yml文件
├── det_r50_vd_db.yml # 测试server版ppocr检测模型训练的yml文件
├── rec_icdar15_r34_train.yml # 测试server版ppocr识别模型训练的yml文件
├── ppocr_sys_mobile_params.txt # 测试mobile版ppocr检测+识别模型串联的参数配置文件
├── ppocr_det_mobile_params.txt # 测试mobile版ppocr检测模型的参数配置文件
├── ppocr_rec_mobile_params.txt # 测试mobile版ppocr识别模型的参数配置文件
├── ppocr_sys_server_params.txt # 测试server版ppocr检测+识别模型串联的参数配置文件
├── ppocr_det_server_params.txt # 测试server版ppocr检测模型的参数配置文件
├── ppocr_rec_server_params.txt # 测试server版ppocr识别模型的参数配置文件
├── ...
├── results/ # 预先保存的预测结果,用于和实际预测结果进行精读比对
├── ppocr_det_mobile_results_fp32.txt # 预存的mobile版ppocr检测模型fp32精度的结果
├── ppocr_det_mobile_results_fp16.txt # 预存的mobile版ppocr检测模型fp16精度的结果
├── ppocr_det_mobile_results_fp32_cpp.txt # 预存的mobile版ppocr检测模型c++预测的fp32精度的结果
├── ppocr_det_mobile_results_fp16_cpp.txt # 预存的mobile版ppocr检测模型c++预测的fp16精度的结果
├── ...
├── prepare.sh # 完成test_*.sh运行所需要的数据和模型下载
├── test_python.sh # 测试python训练预测的主程序
├── test_cpp.sh # 测试c++预测的主程序
├── test_serving.sh # 测试serving部署预测的主程序
├── test_lite.sh # 测试lite部署预测的主程序
├── compare_results.py # 用于对比log中的预测结果与results中的预存结果精度误差是否在限定范围内
└── readme.md # 使用文档
```
- 模式4:whole_train_infer , CE: 全量数据训练,全量数据预测,验证模型训练精度,预测精度,预测速度;
```shell
bash tests/prepare.sh ./tests/ocr_det_params.txt 'whole_train_infer'
bash tests/test.sh ./tests/ocr_det_params.txt 'whole_train_infer'
```
### 测试流程
使用本工具,可以测试不同功能的支持情况,以及预测结果是否对齐,测试流程如下:
<div align="center">
<img src="docs/test.png" width="800">
</div>
- 模式5:cpp_infer , CE: 验证inference model的c++预测是否走通;
```shell
bash tests/prepare.sh ./tests/ocr_det_params.txt 'cpp_infer'
bash tests/test.sh ./tests/ocr_det_params.txt 'cpp_infer'
```
1. 运行prepare.sh准备测试所需数据和模型;
2. 运行要测试的功能对应的测试脚本`test_*.sh`,产出log,由log可以看到不同配置是否运行成功;
3.`compare_results.py`对比log中的预测结果和预存在results目录下的结果,判断预测精度是否符合预期(在误差范围内)。
# 日志输出
最终在```tests/output```目录下生成.log后缀的日志文件
其中,有4个测试主程序,功能如下:
- `test_python.sh`:测试基于Python的模型训练、评估、推理等基本功能,包括裁剪、量化、蒸馏。
- `test_cpp.sh`:测试基于C++的模型推理。
- `test_serving.sh`:测试基于Paddle Serving的服务化部署功能。
- `test_lite.sh`:测试基于Paddle-Lite的端侧预测部署功能。
各功能测试中涉及GPU/CPU、mkldnn、Tensorrt等多种参数配置,点击相应链接了解更多细节和使用教程:
[test_python使用](docs/test_python.md)
[test_cpp使用](docs/test_cpp.md)
[test_serving使用](docs/test_serving.md)
[test_lite使用](docs/test_lite.md)
This diff is collapsed.
......@@ -2,7 +2,14 @@
source tests/common_func.sh
FILENAME=$1
dataline=$(awk 'NR==1, NR==51{print}' $FILENAME)
# MODE be one of ['lite_train_infer' 'whole_infer' 'whole_train_infer', 'infer', 'klquant_infer']
MODE=$2
if [ ${MODE} = "klquant_infer" ]; then
dataline=$(awk 'NR==82, NR==98{print}' $FILENAME)
else
dataline=$(awk 'NR==1, NR==51{print}' $FILENAME)
fi
# parser params
IFS=$'\n'
......@@ -84,6 +91,35 @@ benchmark_value=$(func_parser_value "${lines[49]}")
infer_key1=$(func_parser_key "${lines[50]}")
infer_value1=$(func_parser_value "${lines[50]}")
# parser klquant_infer
if [ ${MODE} = "klquant_infer" ]; then
# parser inference model
infer_model_dir_list=$(func_parser_value "${lines[1]}")
infer_export_list=$(func_parser_value "${lines[2]}")
infer_is_quant=$(func_parser_value "${lines[3]}")
# parser inference
inference_py=$(func_parser_value "${lines[4]}")
use_gpu_key=$(func_parser_key "${lines[5]}")
use_gpu_list=$(func_parser_value "${lines[5]}")
use_mkldnn_key=$(func_parser_key "${lines[6]}")
use_mkldnn_list=$(func_parser_value "${lines[6]}")
cpu_threads_key=$(func_parser_key "${lines[7]}")
cpu_threads_list=$(func_parser_value "${lines[7]}")
batch_size_key=$(func_parser_key "${lines[8]}")
batch_size_list=$(func_parser_value "${lines[8]}")
use_trt_key=$(func_parser_key "${lines[9]}")
use_trt_list=$(func_parser_value "${lines[9]}")
precision_key=$(func_parser_key "${lines[10]}")
precision_list=$(func_parser_value "${lines[10]}")
infer_model_key=$(func_parser_key "${lines[11]}")
image_dir_key=$(func_parser_key "${lines[12]}")
infer_img_dir=$(func_parser_value "${lines[12]}")
save_log_key=$(func_parser_key "${lines[13]}")
benchmark_key=$(func_parser_key "${lines[14]}")
benchmark_value=$(func_parser_value "${lines[14]}")
infer_key1=$(func_parser_key "${lines[15]}")
infer_value1=$(func_parser_value "${lines[15]}")
fi
LOG_PATH="./tests/output"
mkdir -p ${LOG_PATH}
......@@ -107,7 +143,7 @@ function func_inference(){
fi
for threads in ${cpu_threads_list[*]}; do
for batch_size in ${batch_size_list[*]}; do
_save_log_path="${_log_path}/infer_cpu_usemkldnn_${use_mkldnn}_threads_${threads}_batchsize_${batch_size}.log"
_save_log_path="${_log_path}/python_infer_cpu_usemkldnn_${use_mkldnn}_threads_${threads}_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}")
......@@ -135,7 +171,7 @@ function func_inference(){
continue
fi
for batch_size in ${batch_size_list[*]}; do
_save_log_path="${_log_path}/infer_gpu_usetrt_${use_trt}_precision_${precision}_batchsize_${batch_size}.log"
_save_log_path="${_log_path}/python_infer_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}")
......@@ -158,16 +194,148 @@ function func_inference(){
done
}
# set cuda device
GPUID=$2
if [ ${#GPUID} -le 0 ];then
env=" "
if [ ${MODE} = "infer" ] || [ ${MODE} = "klquant_infer" ]; then
GPUID=$3
if [ ${#GPUID} -le 0 ];then
env=" "
else
env="export CUDA_VISIBLE_DEVICES=${GPUID}"
fi
# set CUDA_VISIBLE_DEVICES
eval $env
export Count=0
IFS="|"
infer_run_exports=(${infer_export_list})
infer_quant_flag=(${infer_is_quant})
for infer_model in ${infer_model_dir_list[*]}; do
# run export
if [ ${infer_run_exports[Count]} != "null" ];then
save_infer_dir=$(dirname $infer_model)
set_export_weight=$(func_set_params "${export_weight}" "${infer_model}")
set_save_infer_key=$(func_set_params "${save_infer_key}" "${save_infer_dir}")
export_cmd="${python} ${infer_run_exports[Count]} ${set_export_weight} ${set_save_infer_key}"
echo ${infer_run_exports[Count]}
echo $export_cmd
eval $export_cmd
status_export=$?
status_check $status_export "${export_cmd}" "${status_log}"
else
save_infer_dir=${infer_model}
fi
#run inference
is_quant=${infer_quant_flag[Count]}
func_inference "${python}" "${inference_py}" "${save_infer_dir}" "${LOG_PATH}" "${infer_img_dir}" ${is_quant}
Count=$(($Count + 1))
done
else
env="export CUDA_VISIBLE_DEVICES=${GPUID}"
fi
set CUDA_VISIBLE_DEVICES
eval $env
IFS="|"
export Count=0
USE_GPU_KEY=(${train_use_gpu_value})
for gpu in ${gpu_list[*]}; do
use_gpu=${USE_GPU_KEY[Count]}
Count=$(($Count + 1))
if [ ${gpu} = "-1" ];then
env=""
elif [ ${#gpu} -le 1 ];then
env="export CUDA_VISIBLE_DEVICES=${gpu}"
eval ${env}
elif [ ${#gpu} -le 15 ];then
IFS=","
array=(${gpu})
env="export CUDA_VISIBLE_DEVICES=${array[0]}"
IFS="|"
else
IFS=";"
array=(${gpu})
ips=${array[0]}
gpu=${array[1]}
IFS="|"
env=" "
fi
for autocast in ${autocast_list[*]}; do
for trainer in ${trainer_list[*]}; do
flag_quant=False
if [ ${trainer} = ${pact_key} ]; then
run_train=${pact_trainer}
run_export=${pact_export}
flag_quant=True
elif [ ${trainer} = "${fpgm_key}" ]; then
run_train=${fpgm_trainer}
run_export=${fpgm_export}
elif [ ${trainer} = "${distill_key}" ]; then
run_train=${distill_trainer}
run_export=${distill_export}
elif [ ${trainer} = ${trainer_key1} ]; then
run_train=${trainer_value1}
run_export=${export_value1}
elif [[ ${trainer} = ${trainer_key2} ]]; then
run_train=${trainer_value2}
run_export=${export_value2}
else
run_train=${norm_trainer}
run_export=${norm_export}
fi
if [ ${run_train} = "null" ]; then
continue
fi
set_autocast=$(func_set_params "${autocast_key}" "${autocast}")
set_epoch=$(func_set_params "${epoch_key}" "${epoch_num}")
set_pretrain=$(func_set_params "${pretrain_model_key}" "${pretrain_model_value}")
set_batchsize=$(func_set_params "${train_batch_key}" "${train_batch_value}")
set_train_params1=$(func_set_params "${train_param_key1}" "${train_param_value1}")
set_use_gpu=$(func_set_params "${train_use_gpu_key}" "${use_gpu}")
save_log="${LOG_PATH}/${trainer}_gpus_${gpu}_autocast_${autocast}"
# load pretrain from norm training if current trainer is pact or fpgm trainer
if [ ${trainer} = ${pact_key} ] || [ ${trainer} = ${fpgm_key} ]; then
set_pretrain="${load_norm_train_model}"
fi
set_save_model=$(func_set_params "${save_model_key}" "${save_log}")
if [ ${#gpu} -le 2 ];then # train with cpu or single gpu
cmd="${python} ${run_train} ${set_use_gpu} ${set_save_model} ${set_epoch} ${set_pretrain} ${set_autocast} ${set_batchsize} ${set_train_params1} "
elif [ ${#gpu} -le 15 ];then # train with multi-gpu
cmd="${python} -m paddle.distributed.launch --gpus=${gpu} ${run_train} ${set_save_model} ${set_epoch} ${set_pretrain} ${set_autocast} ${set_batchsize} ${set_train_params1}"
else # train with multi-machine
cmd="${python} -m paddle.distributed.launch --ips=${ips} --gpus=${gpu} ${run_train} ${set_save_model} ${set_pretrain} ${set_epoch} ${set_autocast} ${set_batchsize} ${set_train_params1}"
fi
# run train
eval "unset CUDA_VISIBLE_DEVICES"
eval $cmd
status_check $? "${cmd}" "${status_log}"
set_eval_pretrain=$(func_set_params "${pretrain_model_key}" "${save_log}/${train_model_name}")
# save norm trained models to set pretrain for pact training and fpgm training
if [ ${trainer} = ${trainer_norm} ]; then
load_norm_train_model=${set_eval_pretrain}
fi
# run eval
if [ ${eval_py} != "null" ]; then
set_eval_params1=$(func_set_params "${eval_key1}" "${eval_value1}")
eval_cmd="${python} ${eval_py} ${set_eval_pretrain} ${set_use_gpu} ${set_eval_params1}"
eval $eval_cmd
status_check $? "${eval_cmd}" "${status_log}"
fi
# run export model
if [ ${run_export} != "null" ]; then
# run export model
save_infer_path="${save_log}"
set_export_weight=$(func_set_params "${export_weight}" "${save_log}/${train_model_name}")
set_save_infer_key=$(func_set_params "${save_infer_key}" "${save_infer_path}")
export_cmd="${python} ${run_export} ${set_export_weight} ${set_save_infer_key}"
eval $export_cmd
status_check $? "${export_cmd}" "${status_log}"
#run inference
eval $env
save_infer_path="${save_log}"
func_inference "${python}" "${inference_py}" "${save_infer_path}" "${LOG_PATH}" "${train_infer_img_dir}" "${flag_quant}"
eval "unset CUDA_VISIBLE_DEVICES"
fi
done # done with: for trainer in ${trainer_list[*]}; do
done # done with: for autocast in ${autocast_list[*]}; do
done # done with: for gpu in ${gpu_list[*]}; do
fi # end if [ ${MODE} = "infer" ]; then
echo "################### run test ###################"
......@@ -141,7 +141,6 @@ if __name__ == "__main__":
img, flag = check_and_read_gif(image_file)
if not flag:
img = cv2.imread(image_file)
img = img[:, :, ::-1]
if img is None:
logger.info("error in loading image:{}".format(image_file))
continue
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment