Skip to content

P
PaddleClas

PaddlePaddle / PaddleClas
1 年多 前同步成功

通知 116
Star 4999
Fork 1114
P
PaddleClas
提交 4aa32f1f 编写于 作者: D dongshuilong
浏览文件
操作

update docs and add shell scripts

上级 62b3775e
隐藏空白更改
内联 并排
Showing with 841 addition and 15 deletion
test_tipc/README.md
浏览文件 @ 4aa32f1f
......@@ -24,15 +24,15 @@
| :--- | :--- | :----: | :--------: | :---- | :---- | :---- |
| ResNet |ResNet50_vd | 分类 | 支持 | 多机多卡 <br> 混合精度 | FPGM裁剪 <br> PACT量化| |
| MobileNetV3 |MobileNetV3_large_x1_0 | 分类 | 支持 | 多机多卡 <br> 混合精度 | FPGM裁剪 <br> PACT量化| |
| PPLCNet |PPLCNet_x2_5 | 分类 | 支持 | 多机多卡 <br> 混合精度 | FPGM裁剪 <br> PACT量化| |
| PPLCNet |PPLCNet_x2_5 | 分类 | 支持 | 多机多卡 <br> 混合精度 | - | |
## 3. 一键测试工具使用
## 3. 测试工具简介
### 目录介绍
```
./test_tipc/
├── common_func.sh #test_*.sh会调用到的公共函数
├── config # 配置文件目录
│ ├── MobileNetV3_large_x1_0 # MobileNetV3系列模型测试配置文件目录
│ ├── MobileNetV3 # MobileNetV3系列模型测试配置文件目录
│ │ ├── MobileNetV3_large_x1_0_train_infer_python.txt #基础训练预测配置文件
│ │ ├── MobileNetV3_large_x1_0_train_linux_gpu_fleet_amp_infer_python_linux_gpu_cpu.txt #多机多卡训练预测配置文件
│ │ └── MobileNetV3_large_x1_0_train_linux_gpu_normal_amp_infer_python_linux_gpu_cpu.txt #混合精度训练预测配置文件
......@@ -50,7 +50,7 @@
└── test_train_inference_python.sh # 测试python训练预测的主程序
```
### 测试流程
### 测试流程概述
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">
......@@ -60,16 +60,49 @@
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. 用`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）。
其中，有4个测试主程序，功能如下：
- `test_train_inference_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_inference_cpp.sh`：测试基于C++的模型推理。待支持
- `test_serving.sh`：测试基于Paddle Serving的服务化部署功能。待支持
- `test_lite.sh`：测试基于Paddle-Lite的端侧预测部署功能。待支持
测试单项功能仅需两行命令，**如需测试不同模型/功能，替换配置文件即可**，命令格式如下：
```shell
# 功能：准备数据
# 格式：bash + 运行脚本 + 参数1: 配置文件选择 + 参数2: 模式选择
bash test_tipc/prepare.sh configs/[model_name]/[params_file_name] [Mode]
# 功能：运行测试
# 格式：bash + 运行脚本 + 参数1: 配置文件选择 + 参数2: 模式选择
bash test_tipc/test_train_inference_python.sh configs/[model_name]/[params_file_name] [Mode]
```
例如，测试基本训练预测功能的`lite_train_lite_infer`模式，运行：
```shell
# 准备数据
bash test_tipc/prepare.sh ./test_tipc/configs/MobileNetV3/MobileNetV3_large_x1_0_train_infer_python.txt 'lite_train_lite_infer'
# 运行测试
bash test_tipc/test_train_inference_python.sh ./test_tipc/configs/MobileNetV3/MobileNetV3_large_x1_0_train_infer_python.txt 'lite_train_lite_infer'
```
关于本示例命令的更多信息可查看[基础训练预测使用文档](docs/test_train_inference_python.md)。
### 配置文件命名规范
在`configs`目录下，**按模型系列划分为子目录**，子目录中存放所有该模型系列测试需要用到的配置文件，如`MobileNetV3`文件夹下存放了所有`MobileNetV3`系列模型的配置文件。配置文件的命名遵循如下规范：
1. 基础训练预测配置简单命名为：`ModelName_train_infer_python.txt`，表示**Linux环境下单机、不使用混合精度训练+python预测**，其完整命名对应`ModelName_train_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt`，由于本配置文件使用频率较高，这里进行了名称简化。其中`ModelName`指具体模型名称
2. 其他带训练配置命名格式为：`ModelName_train_训练硬件环境(linux_gpu/linux_dcu/…)_是否多机(fleet/normal)_是否混合精度(amp/normal)_预测模式(infer/lite/serving/js)_语言(cpp/python/java)_预测硬件环境(ModelName_linux_gpu/mac/jetson/opencl_arm_gpu/...).txt`。如，linux gpu下多机多卡+混合精度链条测试对应配置 `ModelName_train_linux_gpu_fleet_amp_infer_python_linux_gpu_cpu.txt`，linux dcu下基础训练预测对应配置 `ModelName_train_linux_dcu_normal_normal_infer_python_linux_dcu.txt`。
3. 仅预测的配置（如serving、lite等）命名格式：`ModelName_model_训练硬件环境(ModelName_linux_gpu/linux_dcu/…)_是否多机(fleet/normal)_是否混合精度(amp/normal)_(infer/lite/serving/js)_语言(cpp/python/java)_预测硬件环境(ModelName_linux_gpu/mac/jetson/opencl_arm_gpu/...).txt`，即，与2相比，仅第二个字段从train换为model，测试时模型直接下载获取，这里的“训练硬件环境”表示所测试的模型是在哪种环境下训练得到的。
**根据上述命名规范，可以直接从子目录名称和配置文件名找到需要测试的场景和功能对应的配置文件。**
<a name="more"></a>
#### 更多教程
## 4 开始测试
各功能测试中涉及混合精度、裁剪、量化等训练相关，及mkldnn、Tensorrt等多种预测相关参数配置，请点击下方相应链接了解更多细节和使用教程：
[test_train_inference_python 使用](docs/test_train_inference_python.md)
[test_inference_cpp 使用](docs/test_inference_cpp.md)
[test_serving 使用](docs/test_serving.md)
[test_lite 使用](docs/test_lite.md)
- [test_train_inference_python 使用](docs/test_train_inference_python.md)：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_inference_cpp 使用](docs/test_inference_cpp.md) ：测试基于C++的模型推理。
- [test_serving 使用](docs/test_serving.md) ：测试基于Paddle Serving的服务化部署功能。
- [test_lite_arm_cpu_cpp 使用](docs/test_lite_arm_cpu_cpp.md): 测试基于Paddle-Lite的ARM CPU端c++预测部署功能.
- [test_paddle2onnx 使用](docs/test_paddle2onnx.md)：测试Paddle2ONNX的模型转化功能，并验证正确性。
test_tipc/prepare.sh
浏览文件 @ 4aa32f1f
......@@ -76,7 +76,7 @@ elif [ ${MODE} = "whole_train_whole_infer" ];then
ln -s whole_chain_CIFAR100 ILSVRC2012
cd ILSVRC2012
mv train.txt train_list.txt
mv val.txt val_list.txt
mv test.txt val_list.txt
cd ../../
fi
......
test_tipc/prepare_lite_cpp.sh 0 → 100644
浏览文件 @ 4aa32f1f
#!/bin/bash
source ./test_tipc/common_func.sh
FILENAME=$1
dataline=$(cat ${FILENAME})
# parser params
IFS=$'\n'
lines=(${dataline})
IFS=$'\n'
inference_cmd=$(func_parser_value "${lines[1]}")
DEVICE=$(func_parser_value "${lines[2]}")
det_lite_model_list=$(func_parser_value "${lines[3]}")
rec_lite_model_list=$(func_parser_value "${lines[4]}")
cls_lite_model_list=$(func_parser_value "${lines[5]}")
if [[ $inference_cmd =~ "det" ]];then
lite_model_list=${det_lite_model_list}
elif [[ $inference_cmd =~ "rec" ]];then
lite_model_list=(${rec_lite_model_list[*]} ${cls_lite_model_list[*]})
elif [[ $inference_cmd =~ "system" ]];then
lite_model_list=(${det_lite_model_list[*]} ${rec_lite_model_list[*]} ${cls_lite_model_list[*]})
else
echo "inference_cmd is wrong, please check."
exit 1
fi
if [ ${DEVICE} = "ARM_CPU" ];then
valid_targets="arm"
paddlelite_url="https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.10-rc/inference_lite_lib.android.armv8.gcc.c++_shared.with_extra.with_cv.tar.gz"
end_index="66"
elif [ ${DEVICE} = "ARM_GPU_OPENCL" ];then
valid_targets="opencl"
paddlelite_url="https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.10-rc/inference_lite_lib.armv8.clang.with_exception.with_extra.with_cv.opencl.tar.gz"
end_index="71"
else
echo "DEVICE only suport ARM_CPU, ARM_GPU_OPENCL."
exit 2
fi
# prepare lite .nb model
pip install paddlelite==2.10-rc
current_dir=${PWD}
IFS="|"
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 [[ $model =~ "v2.0" ]];then
inference_model_url=https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/${model}.tar
else
echo "Model is wrong, please check."
exit 3
fi
inference_model=${inference_model_url##*/}
wget -nc -P ${model_path} ${inference_model_url}
cd ${model_path} && tar -xf ${inference_model} && cd ../
model_dir=${model_path}/${inference_model%.*}
model_file=${model_dir}/inference.pdmodel
param_file=${model_dir}/inference.pdiparams
paddle_lite_opt --model_dir=${model_dir} --model_file=${model_file} --param_file=${param_file} --valid_targets=${valid_targets} --optimize_out=${model_dir}_opt
done
# prepare test data
data_url=https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/icdar2015_lite.tar
model_path=./inference_models
inference_model=${inference_model_url##*/}
data_file=${data_url##*/}
wget -nc -P ./inference_models ${inference_model_url}
wget -nc -P ./test_data ${data_url}
cd ./inference_models && tar -xf ${inference_model} && cd ../
cd ./test_data && tar -xf ${data_file} && rm ${data_file} && cd ../
# prepare lite env
paddlelite_zipfile=$(echo $paddlelite_url | awk -F "/" '{print $NF}')
paddlelite_file=${paddlelite_zipfile:0:${end_index}}
wget ${paddlelite_url} && tar -xf ${paddlelite_zipfile}
mkdir -p ${paddlelite_file}/demo/cxx/ocr/test_lite
cp -r ${model_path}/*_opt.nb test_data ${paddlelite_file}/demo/cxx/ocr/test_lite
cp ppocr/utils/ppocr_keys_v1.txt deploy/lite/config.txt ${paddlelite_file}/demo/cxx/ocr/test_lite
cp -r ./deploy/lite/* ${paddlelite_file}/demo/cxx/ocr/
cp ${paddlelite_file}/cxx/lib/libpaddle_light_api_shared.so ${paddlelite_file}/demo/cxx/ocr/test_lite
cp ${FILENAME} test_tipc/test_lite_arm_cpp.sh test_tipc/common_func.sh ${paddlelite_file}/demo/cxx/ocr/test_lite
cd ${paddlelite_file}/demo/cxx/ocr/
git clone https://github.com/cuicheng01/AutoLog.git
# make
make -j
sleep 1
make -j
cp ocr_db_crnn test_lite && cp test_lite/libpaddle_light_api_shared.so test_lite/libc++_shared.so
tar -cf test_lite.tar ./test_lite && cp test_lite.tar ${current_dir} && cd ${current_dir}
rm -rf ${paddlelite_file}* && rm -rf ${model_path}
test_tipc/test_inference_cpp.sh 0 → 100644
浏览文件 @ 4aa32f1f
#!/bin/bash
source test_tipc/common_func.sh
FILENAME=$1
dataline=$(awk 'NR==1, NR==16{print}' $FILENAME)
# parser params
IFS=$'\n'
lines=(${dataline})
# parser cpp inference model
model_name=$(func_parser_value "${lines[1]}")
use_opencv=$(func_parser_value "${lines[2]}")
cpp_infer_model_dir_list=$(func_parser_value "${lines[3]}")
cpp_infer_is_quant=$(func_parser_value "${lines[4]}")
# parser cpp inference
inference_cmd=$(func_parser_value "${lines[5]}")
cpp_use_gpu_key=$(func_parser_key "${lines[6]}")
cpp_use_gpu_list=$(func_parser_value "${lines[6]}")
cpp_use_mkldnn_key=$(func_parser_key "${lines[7]}")
cpp_use_mkldnn_list=$(func_parser_value "${lines[7]}")
cpp_cpu_threads_key=$(func_parser_key "${lines[8]}")
cpp_cpu_threads_list=$(func_parser_value "${lines[8]}")
cpp_batch_size_key=$(func_parser_key "${lines[9]}")
cpp_batch_size_list=$(func_parser_value "${lines[9]}")
cpp_use_trt_key=$(func_parser_key "${lines[10]}")
cpp_use_trt_list=$(func_parser_value "${lines[10]}")
cpp_precision_key=$(func_parser_key "${lines[11]}")
cpp_precision_list=$(func_parser_value "${lines[11]}")
cpp_infer_model_key=$(func_parser_key "${lines[12]}")
cpp_image_dir_key=$(func_parser_key "${lines[13]}")
cpp_infer_img_dir=$(func_parser_value "${lines[13]}")
cpp_infer_key1=$(func_parser_key "${lines[14]}")
cpp_infer_value1=$(func_parser_value "${lines[14]}")
cpp_benchmark_key=$(func_parser_key "${lines[15]}")
cpp_benchmark_value=$(func_parser_value "${lines[15]}")
LOG_PATH="./test_tipc/output"
mkdir -p ${LOG_PATH}
status_log="${LOG_PATH}/results_cpp.log"
function func_cpp_inference(){
IFS='|'
_script=$1
_model_dir=$2
_log_path=$3
_img_dir=$4
_flag_quant=$5
# inference
for use_gpu in ${cpp_use_gpu_list[*]}; do
if [ ${use_gpu} = "False" ] || [ ${use_gpu} = "cpu" ]; then
for use_mkldnn in ${cpp_use_mkldnn_list[*]}; do
if [ ${use_mkldnn} = "False" ] && [ ${_flag_quant} = "True" ]; then
continue
fi
for threads in ${cpp_cpu_threads_list[*]}; do
for batch_size in ${cpp_batch_size_list[*]}; do
precision="fp32"
if [ ${use_mkldnn} = "False" ] && [ ${_flag_quant} = "True" ]; then
precison="int8"
fi
_save_log_path="${_log_path}/cpp_infer_cpu_usemkldnn_${use_mkldnn}_threads_${threads}_precision_${precision}_batchsize_${batch_size}.log"
set_infer_data=$(func_set_params "${cpp_image_dir_key}" "${_img_dir}")
set_benchmark=$(func_set_params "${cpp_benchmark_key}" "${cpp_benchmark_value}")
set_batchsize=$(func_set_params "${cpp_batch_size_key}" "${batch_size}")
set_cpu_threads=$(func_set_params "${cpp_cpu_threads_key}" "${threads}")
set_model_dir=$(func_set_params "${cpp_infer_model_key}" "${_model_dir}")
set_infer_params1=$(func_set_params "${cpp_infer_key1}" "${cpp_infer_value1}")
command="${_script} ${cpp_use_gpu_key}=${use_gpu} ${cpp_use_mkldnn_key}=${use_mkldnn} ${set_cpu_threads} ${set_model_dir} ${set_batchsize} ${set_infer_data} ${set_benchmark} ${set_infer_params1} > ${_save_log_path} 2>&1 "
eval $command
last_status=${PIPESTATUS[0]}
eval "cat ${_save_log_path}"
status_check $last_status "${command}" "${status_log}"
done
done
done
elif [ ${use_gpu} = "True" ] || [ ${use_gpu} = "gpu" ]; then
for use_trt in ${cpp_use_trt_list[*]}; do
for precision in ${cpp_precision_list[*]}; do
if [[ ${_flag_quant} = "False" ]] && [[ ${precision} =~ "int8" ]]; then
continue
fi
if [[ ${precision} =~ "fp16" || ${precision} =~ "int8" ]] && [ ${use_trt} = "False" ]; then
continue
fi
if [[ ${use_trt} = "False" || ${precision} =~ "int8" ]] && [ ${_flag_quant} = "True" ]; then
continue
fi
for batch_size in ${cpp_batch_size_list[*]}; do
_save_log_path="${_log_path}/cpp_infer_gpu_usetrt_${use_trt}_precision_${precision}_batchsize_${batch_size}.log"
set_infer_data=$(func_set_params "${cpp_image_dir_key}" "${_img_dir}")
set_benchmark=$(func_set_params "${cpp_benchmark_key}" "${cpp_benchmark_value}")
set_batchsize=$(func_set_params "${cpp_batch_size_key}" "${batch_size}")
set_tensorrt=$(func_set_params "${cpp_use_trt_key}" "${use_trt}")
set_precision=$(func_set_params "${cpp_precision_key}" "${precision}")
set_model_dir=$(func_set_params "${cpp_infer_model_key}" "${_model_dir}")
set_infer_params1=$(func_set_params "${cpp_infer_key1}" "${cpp_infer_value1}")
command="${_script} ${cpp_use_gpu_key}=${use_gpu} ${set_tensorrt} ${set_precision} ${set_model_dir} ${set_batchsize} ${set_infer_data} ${set_benchmark} ${set_infer_params1} > ${_save_log_path} 2>&1 "
eval $command
last_status=${PIPESTATUS[0]}
eval "cat ${_save_log_path}"
status_check $last_status "${command}" "${status_log}"
done
done
done
else
echo "Does not support hardware other than CPU and GPU Currently!"
fi
done
}
cd deploy/cpp_infer
if [ ${use_opencv} = "True" ]; then
if [ -d "opencv-3.4.7/opencv3/" ] && [ $(md5sum opencv-3.4.7.tar.gz | awk -F ' ' '{print $1}') = "faa2b5950f8bee3f03118e600c74746a" ];then
echo "################### build opencv skipped ###################"
else
echo "################### build opencv ###################"
rm -rf opencv-3.4.7.tar.gz opencv-3.4.7/
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/opencv-3.4.7.tar.gz
tar -xf opencv-3.4.7.tar.gz
cd opencv-3.4.7/
install_path=$(pwd)/opencv3
rm -rf build
mkdir build
cd build
cmake .. \
-DCMAKE_INSTALL_PREFIX=${install_path} \
-DCMAKE_BUILD_TYPE=Release \
-DBUILD_SHARED_LIBS=OFF \
-DWITH_IPP=OFF \
-DBUILD_IPP_IW=OFF \
-DWITH_LAPACK=OFF \
-DWITH_EIGEN=OFF \
-DCMAKE_INSTALL_LIBDIR=lib64 \
-DWITH_ZLIB=ON \
-DBUILD_ZLIB=ON \
-DWITH_JPEG=ON \
-DBUILD_JPEG=ON \
-DWITH_PNG=ON \
-DBUILD_PNG=ON \
-DWITH_TIFF=ON \
-DBUILD_TIFF=ON
make -j
make install
cd ../
echo "################### build opencv finished ###################"
fi
fi
echo "################### build PaddleOCR demo ####################"
if [ ${use_opencv} = "True" ]; then
OPENCV_DIR=$(pwd)/opencv-3.4.7/opencv3/
else
OPENCV_DIR=''
fi
LIB_DIR=$(pwd)/Paddle/build/paddle_inference_install_dir/
CUDA_LIB_DIR=$(dirname `find /usr -name libcudart.so`)
CUDNN_LIB_DIR=$(dirname `find /usr -name libcudnn.so`)
BUILD_DIR=build
rm -rf ${BUILD_DIR}
mkdir ${BUILD_DIR}
cd ${BUILD_DIR}
cmake .. \
-DPADDLE_LIB=${LIB_DIR} \
-DWITH_MKL=ON \
-DWITH_GPU=OFF \
-DWITH_STATIC_LIB=OFF \
-DWITH_TENSORRT=OFF \
-DOPENCV_DIR=${OPENCV_DIR} \
-DCUDNN_LIB=${CUDNN_LIB_DIR} \
-DCUDA_LIB=${CUDA_LIB_DIR} \
-DTENSORRT_DIR=${TENSORRT_DIR} \
make -j
cd ../../../
echo "################### build PaddleOCR demo finished ###################"
# set cuda device
GPUID=$2
if [ ${#GPUID} -le 0 ];then
env=" "
else
env="export CUDA_VISIBLE_DEVICES=${GPUID}"
fi
set CUDA_VISIBLE_DEVICES
eval $env
echo "################### run test ###################"
export Count=0
IFS="|"
infer_quant_flag=(${cpp_infer_is_quant})
for infer_model in ${cpp_infer_model_dir_list[*]}; do
#run inference
is_quant=${infer_quant_flag[Count]}
func_cpp_inference "${inference_cmd}" "${infer_model}" "${LOG_PATH}" "${cpp_infer_img_dir}" ${is_quant}
Count=$(($Count + 1))
done
test_tipc/test_inference_jeston.sh 0 → 100644
浏览文件 @ 4aa32f1f
#!/bin/bash
source test_tipc/common_func.sh
source test_tipc/test_train_inference_python.sh
FILENAME=$1
# MODE be one of ['whole_infer']
MODE=$2
dataline=$(awk 'NR==1, NR==17{print}' $FILENAME)
# parser params
IFS=$'\n'
lines=(${dataline})
model_name=$(func_parser_value "${lines[1]}")
python=$(func_parser_value "${lines[2]}")
infer_model_dir_list=$(func_parser_value "${lines[3]}")
infer_export_list=$(func_parser_value "${lines[4]}")
infer_is_quant=$(func_parser_value "${lines[5]}")
# parser inference
inference_py=$(func_parser_value "${lines[6]}")
use_gpu_key=$(func_parser_key "${lines[7]}")
use_gpu_list=$(func_parser_value "${lines[7]}")
use_mkldnn_key=$(func_parser_key "${lines[8]}")
use_mkldnn_list=$(func_parser_value "${lines[8]}")
cpu_threads_key=$(func_parser_key "${lines[9]}")
cpu_threads_list=$(func_parser_value "${lines[9]}")
batch_size_key=$(func_parser_key "${lines[10]}")
batch_size_list=$(func_parser_value "${lines[10]}")
use_trt_key=$(func_parser_key "${lines[11]}")
use_trt_list=$(func_parser_value "${lines[11]}")
precision_key=$(func_parser_key "${lines[12]}")
precision_list=$(func_parser_value "${lines[12]}")
infer_model_key=$(func_parser_key "${lines[13]}")
image_dir_key=$(func_parser_key "${lines[14]}")
infer_img_dir=$(func_parser_value "${lines[14]}")
save_log_key=$(func_parser_key "${lines[15]}")
benchmark_key=$(func_parser_key "${lines[16]}")
benchmark_value=$(func_parser_value "${lines[16]}")
infer_key1=$(func_parser_key "${lines[17]}")
infer_value1=$(func_parser_value "${lines[17]}")
LOG_PATH="./test_tipc/output"
mkdir -p ${LOG_PATH}
status_log="${LOG_PATH}/results_python.log"
if [ ${MODE} = "whole_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]}
if [ ${MODE} = "klquant_infer" ]; then
is_quant="True"
fi
func_inference "${python}" "${inference_py}" "${save_infer_dir}" "${LOG_PATH}" "${infer_img_dir}" ${is_quant}
Count=$(($Count + 1))
done
fi
test_tipc/test_lite_arm_cpp.sh 0 → 100644
浏览文件 @ 4aa32f1f
#!/bin/bash
source ./common_func.sh
export LD_LIBRARY_PATH=${PWD}:$LD_LIBRARY_PATH
FILENAME=$1
dataline=$(cat $FILENAME)
# parser params
IFS=$'\n'
lines=(${dataline})
# parser lite inference
inference_cmd=$(func_parser_value "${lines[1]}")
runtime_device=$(func_parser_value "${lines[2]}")
det_model_list=$(func_parser_value "${lines[3]}")
rec_model_list=$(func_parser_value "${lines[4]}")
cls_model_list=$(func_parser_value "${lines[5]}")
cpu_threads_list=$(func_parser_value "${lines[6]}")
det_batch_size_list=$(func_parser_value "${lines[7]}")
rec_batch_size_list=$(func_parser_value "${lines[8]}")
infer_img_dir_list=$(func_parser_value "${lines[9]}")
config_dir=$(func_parser_value "${lines[10]}")
rec_dict_dir=$(func_parser_value "${lines[11]}")
benchmark_value=$(func_parser_value "${lines[12]}")
if [[ $inference_cmd =~ "det" ]]; then
lite_model_list=${det_lite_model_list}
elif [[ $inference_cmd =~ "rec" ]]; then
lite_model_list=(${rec_lite_model_list[*]} ${cls_lite_model_list[*]})
elif [[ $inference_cmd =~ "system" ]]; then
lite_model_list=(${det_lite_model_list[*]} ${rec_lite_model_list[*]} ${cls_lite_model_list[*]})
else
echo "inference_cmd is wrong, please check."
exit 1
fi
LOG_PATH="./output"
mkdir -p ${LOG_PATH}
status_log="${LOG_PATH}/results.log"
function func_test_det(){
IFS='|'
_script=$1
_det_model=$2
_log_path=$3
_img_dir=$4
_config=$5
if [[ $_det_model =~ "slim" ]]; then
precision="INT8"
else
precision="FP32"
fi
# lite inference
for num_threads in ${cpu_threads_list[*]}; do
for det_batchsize in ${det_batch_size_list[*]}; do
_save_log_path="${_log_path}/lite_${_det_model}_runtime_device_${runtime_device}_precision_${precision}_det_batchsize_${det_batchsize}_threads_${num_threads}.log"
command="${_script} ${_det_model} ${runtime_device} ${precision} ${num_threads} ${det_batchsize} ${_img_dir} ${_config} ${benchmark_value} > ${_save_log_path} 2>&1"
eval ${command}
status_check $? "${command}" "${status_log}"
done
done
}
function func_test_rec(){
IFS='|'
_script=$1
_rec_model=$2
_cls_model=$3
_log_path=$4
_img_dir=$5
_config=$6
_rec_dict_dir=$7
if [[ $_det_model =~ "slim" ]]; then
_precision="INT8"
else
_precision="FP32"
fi
# lite inference
for num_threads in ${cpu_threads_list[*]}; do
for rec_batchsize in ${rec_batch_size_list[*]}; do
_save_log_path="${_log_path}/lite_${_rec_model}_${cls_model}_runtime_device_${runtime_device}_precision_${_precision}_rec_batchsize_${rec_batchsize}_threads_${num_threads}.log"
command="${_script} ${_rec_model} ${_cls_model} ${runtime_device} ${_precision} ${num_threads} ${rec_batchsize} ${_img_dir} ${_config} ${_rec_dict_dir} ${benchmark_value} > ${_save_log_path} 2>&1"
eval ${command}
status_check $? "${command}" "${status_log}"
done
done
}
function func_test_system(){
IFS='|'
_script=$1
_det_model=$2
_rec_model=$3
_cls_model=$4
_log_path=$5
_img_dir=$6
_config=$7
_rec_dict_dir=$8
if [[ $_det_model =~ "slim" ]]; then
_precision="INT8"
else
_precision="FP32"
fi
# lite inference
for num_threads in ${cpu_threads_list[*]}; do
for det_batchsize in ${det_batch_size_list[*]}; do
for rec_batchsize in ${rec_batch_size_list[*]}; do
_save_log_path="${_log_path}/lite_${_det_model}_${_rec_model}_${_cls_model}_runtime_device_${runtime_device}_precision_${_precision}_det_batchsize_${det_batchsize}_rec_batchsize_${rec_batchsize}_threads_${num_threads}.log"
command="${_script} ${_det_model} ${_rec_model} ${_cls_model} ${runtime_device} ${_precision} ${num_threads} ${det_batchsize} ${_img_dir} ${_config} ${_rec_dict_dir} ${benchmark_value} > ${_save_log_path} 2>&1"
eval ${command}
status_check $? "${command}" "${status_log}"
done
done
done
}
echo "################### run test ###################"
if [[ $inference_cmd =~ "det" ]]; then
IFS="|"
det_model_list=(${det_model_list[*]})
for i in {0..1}; do
#run lite inference
for img_dir in ${infer_img_dir_list[*]}; do
func_test_det "${inference_cmd}" "${det_model_list[i]}_opt.nb" "${LOG_PATH}" "${img_dir}" "${config_dir}"
done
done
elif [[ $inference_cmd =~ "rec" ]]; then
IFS="|"
rec_model_list=(${rec_model_list[*]})
cls_model_list=(${cls_model_list[*]})
for i in {0..1}; do
#run lite inference
for img_dir in ${infer_img_dir_list[*]}; do
func_test_rec "${inference_cmd}" "${rec_model}_opt.nb" "${cls_model_list[i]}_opt.nb" "${LOG_PATH}" "${img_dir}" "${rec_dict_dir}" "${config_dir}"
done
done
elif [[ $inference_cmd =~ "system" ]]; then
IFS="|"
det_model_list=(${det_model_list[*]})
rec_model_list=(${rec_model_list[*]})
cls_model_list=(${cls_model_list[*]})
for i in {0..1}; do
#run lite inference
for img_dir in ${infer_img_dir_list[*]}; do
func_test_system "${inference_cmd}" "${det_model_list[i]}_opt.nb" "${rec_model_list[i]}_opt.nb" "${cls_model_list[i]}_opt.nb" "${LOG_PATH}" "${img_dir}" "${config_dir}" "${rec_dict_dir}"
done
done
fi
test_tipc/test_paddle2onnx.sh 0 → 100644
浏览文件 @ 4aa32f1f
#!/bin/bash
source test_tipc/common_func.sh
FILENAME=$1
dataline=$(cat ${FILENAME})
lines=(${dataline})
# common params
model_name=$(func_parser_value "${lines[1]}")
python=$(func_parser_value "${lines[2]}")
# parser params
dataline=$(awk 'NR==1, NR==12{print}' $FILENAME)
IFS=$'\n'
lines=(${dataline})
# parser paddle2onnx
model_name=$(func_parser_value "${lines[1]}")
python=$(func_parser_value "${lines[2]}")
padlle2onnx_cmd=$(func_parser_value "${lines[3]}")
infer_model_dir_key=$(func_parser_key "${lines[4]}")
infer_model_dir_value=$(func_parser_value "${lines[4]}")
model_filename_key=$(func_parser_key "${lines[5]}")
model_filename_value=$(func_parser_value "${lines[5]}")
params_filename_key=$(func_parser_key "${lines[6]}")
params_filename_value=$(func_parser_value "${lines[6]}")
save_file_key=$(func_parser_key "${lines[7]}")
save_file_value=$(func_parser_value "${lines[7]}")
opset_version_key=$(func_parser_key "${lines[8]}")
opset_version_value=$(func_parser_value "${lines[8]}")
enable_onnx_checker_key=$(func_parser_key "${lines[9]}")
enable_onnx_checker_value=$(func_parser_value "${lines[9]}")
# parser onnx inference
inference_py=$(func_parser_value "${lines[10]}")
use_gpu_key=$(func_parser_key "${lines[11]}")
use_gpu_value=$(func_parser_value "${lines[11]}")
det_model_key=$(func_parser_key "${lines[12]}")
image_dir_key=$(func_parser_key "${lines[13]}")
image_dir_value=$(func_parser_value "${lines[13]}")
LOG_PATH="./test_tipc/output"
mkdir -p ./test_tipc/output
status_log="${LOG_PATH}/results_paddle2onnx.log"
function func_paddle2onnx(){
IFS='|'
_script=$1
# paddle2onnx
_save_log_path="${LOG_PATH}/paddle2onnx_infer_cpu.log"
set_dirname=$(func_set_params "${infer_model_dir_key}" "${infer_model_dir_value}")
set_model_filename=$(func_set_params "${model_filename_key}" "${model_filename_value}")
set_params_filename=$(func_set_params "${params_filename_key}" "${params_filename_value}")
set_save_model=$(func_set_params "${save_file_key}" "${save_file_value}")
set_opset_version=$(func_set_params "${opset_version_key}" "${opset_version_value}")
set_enable_onnx_checker=$(func_set_params "${enable_onnx_checker_key}" "${enable_onnx_checker_value}")
trans_model_cmd="${padlle2onnx_cmd} ${set_dirname} ${set_model_filename} ${set_params_filename} ${set_save_model} ${set_opset_version} ${set_enable_onnx_checker}"
eval $trans_model_cmd
last_status=${PIPESTATUS[0]}
status_check $last_status "${trans_model_cmd}" "${status_log}"
# python inference
set_gpu=$(func_set_params "${use_gpu_key}" "${use_gpu_value}")
set_model_dir=$(func_set_params "${det_model_key}" "${save_file_value}")
set_img_dir=$(func_set_params "${image_dir_key}" "${image_dir_value}")
infer_model_cmd="${python} ${inference_py} ${set_gpu} ${set_img_dir} ${set_model_dir} --use_onnx=True > ${_save_log_path} 2>&1 "
eval $infer_model_cmd
status_check $last_status "${infer_model_cmd}" "${status_log}"
}
echo "################### run test ###################"
export Count=0
IFS="|"
func_paddle2onnx
\ No newline at end of file
test_tipc/test_serving.sh 0 → 100644
浏览文件 @ 4aa32f1f
#!/bin/bash
source test_tipc/common_func.sh
FILENAME=$1
dataline=$(awk 'NR==1, NR==18{print}' $FILENAME)
# parser params
IFS=$'\n'
lines=(${dataline})
# parser serving
model_name=$(func_parser_value "${lines[1]}")
python=$(func_parser_value "${lines[2]}")
trans_model_py=$(func_parser_value "${lines[3]}")
infer_model_dir_key=$(func_parser_key "${lines[4]}")
infer_model_dir_value=$(func_parser_value "${lines[4]}")
model_filename_key=$(func_parser_key "${lines[5]}")
model_filename_value=$(func_parser_value "${lines[5]}")
params_filename_key=$(func_parser_key "${lines[6]}")
params_filename_value=$(func_parser_value "${lines[6]}")
serving_server_key=$(func_parser_key "${lines[7]}")
serving_server_value=$(func_parser_value "${lines[7]}")
serving_client_key=$(func_parser_key "${lines[8]}")
serving_client_value=$(func_parser_value "${lines[8]}")
serving_dir_value=$(func_parser_value "${lines[9]}")
web_service_py=$(func_parser_value "${lines[10]}")
web_use_gpu_key=$(func_parser_key "${lines[11]}")
web_use_gpu_list=$(func_parser_value "${lines[11]}")
web_use_mkldnn_key=$(func_parser_key "${lines[12]}")
web_use_mkldnn_list=$(func_parser_value "${lines[12]}")
web_cpu_threads_key=$(func_parser_key "${lines[13]}")
web_cpu_threads_list=$(func_parser_value "${lines[13]}")
web_use_trt_key=$(func_parser_key "${lines[14]}")
web_use_trt_list=$(func_parser_value "${lines[14]}")
web_precision_key=$(func_parser_key "${lines[15]}")
web_precision_list=$(func_parser_value "${lines[15]}")
pipeline_py=$(func_parser_value "${lines[16]}")
image_dir_key=$(func_parser_key "${lines[17]}")
image_dir_value=$(func_parser_value "${lines[17]}")
LOG_PATH="../../test_tipc/output"
mkdir -p ./test_tipc/output
status_log="${LOG_PATH}/results_serving.log"
function func_serving(){
IFS='|'
_python=$1
_script=$2
_model_dir=$3
# pdserving
set_dirname=$(func_set_params "${infer_model_dir_key}" "${infer_model_dir_value}")
set_model_filename=$(func_set_params "${model_filename_key}" "${model_filename_value}")
set_params_filename=$(func_set_params "${params_filename_key}" "${params_filename_value}")
set_serving_server=$(func_set_params "${serving_server_key}" "${serving_server_value}")
set_serving_client=$(func_set_params "${serving_client_key}" "${serving_client_value}")
set_image_dir=$(func_set_params "${image_dir_key}" "${image_dir_value}")
trans_model_cmd="${python} ${trans_model_py} ${set_dirname} ${set_model_filename} ${set_params_filename} ${set_serving_server} ${set_serving_client}"
eval $trans_model_cmd
cd ${serving_dir_value}
echo $PWD
unset https_proxy
unset http_proxy
for python in ${python[*]}; do
if [ ${python} = "cpp"]; then
for use_gpu in ${web_use_gpu_list[*]}; do
if [ ${use_gpu} = "null" ]; then
web_service_cpp_cmd="${python} -m paddle_serving_server.serve --model ppocr_det_mobile_2.0_serving/ ppocr_rec_mobile_2.0_serving/ --port 9293"
eval $web_service_cmd
sleep 2s
_save_log_path="${LOG_PATH}/server_infer_cpp_cpu_pipeline_usemkldnn_False_threads_4_batchsize_1.log"
pipeline_cmd="${python} ocr_cpp_client.py ppocr_det_mobile_2.0_client/ ppocr_rec_mobile_2.0_client/"
eval $pipeline_cmd
status_check $last_status "${pipeline_cmd}" "${status_log}"
sleep 2s
ps ux | grep -E 'web_service|pipeline' | awk '{print $2}' | xargs kill -s 9
else
web_service_cpp_cmd="${python} -m paddle_serving_server.serve --model ppocr_det_mobile_2.0_serving/ ppocr_rec_mobile_2.0_serving/ --port 9293 --gpu_id=0"
eval $web_service_cmd
sleep 2s
_save_log_path="${LOG_PATH}/server_infer_cpp_cpu_pipeline_usemkldnn_False_threads_4_batchsize_1.log"
pipeline_cmd="${python} ocr_cpp_client.py ppocr_det_mobile_2.0_client/ ppocr_rec_mobile_2.0_client/"
eval $pipeline_cmd
status_check $last_status "${pipeline_cmd}" "${status_log}"
sleep 2s
ps ux | grep -E 'web_service|pipeline' | awk '{print $2}' | xargs kill -s 9
fi
done
else
# python serving
for use_gpu in ${web_use_gpu_list[*]}; do
echo ${ues_gpu}
if [ ${use_gpu} = "null" ]; then
for use_mkldnn in ${web_use_mkldnn_list[*]}; do
if [ ${use_mkldnn} = "False" ]; then
continue
fi
for threads in ${web_cpu_threads_list[*]}; do
set_cpu_threads=$(func_set_params "${web_cpu_threads_key}" "${threads}")
web_service_cmd="${python} ${web_service_py} ${web_use_gpu_key}=${use_gpu} ${web_use_mkldnn_key}=${use_mkldnn} ${set_cpu_threads} &"
eval $web_service_cmd
sleep 2s
for pipeline in ${pipeline_py[*]}; do
_save_log_path="${LOG_PATH}/server_infer_cpu_${pipeline%_client*}_usemkldnn_${use_mkldnn}_threads_${threads}_batchsize_1.log"
pipeline_cmd="${python} ${pipeline} ${set_image_dir} > ${_save_log_path} 2>&1 "
eval $pipeline_cmd
last_status=${PIPESTATUS[0]}
eval "cat ${_save_log_path}"
status_check $last_status "${pipeline_cmd}" "${status_log}"
sleep 2s
done
ps ux | grep -E 'web_service|pipeline' | awk '{print $2}' | xargs kill -s 9
done
done
elif [ ${use_gpu} = "0" ]; then
for use_trt in ${web_use_trt_list[*]}; do
for precision in ${web_precision_list[*]}; do
if [[ ${_flag_quant} = "False" ]] && [[ ${precision} =~ "int8" ]]; then
continue
fi
if [[ ${precision} =~ "fp16" || ${precision} =~ "int8" ]] && [ ${use_trt} = "False" ]; then
continue
fi
if [[ ${use_trt} = "False" || ${precision} =~ "int8" ]] && [[ ${_flag_quant} = "True" ]]; then
continue
fi
set_tensorrt=$(func_set_params "${web_use_trt_key}" "${use_trt}")
set_precision=$(func_set_params "${web_precision_key}" "${precision}")
web_service_cmd="${python} ${web_service_py} ${web_use_gpu_key}=${use_gpu} ${set_tensorrt} ${set_precision} & "
eval $web_service_cmd
sleep 2s
for pipeline in ${pipeline_py[*]}; do
_save_log_path="${LOG_PATH}/server_infer_gpu_${pipeline%_client*}_usetrt_${use_trt}_precision_${precision}_batchsize_1.log"
pipeline_cmd="${python} ${pipeline} ${set_image_dir}> ${_save_log_path} 2>&1"
eval $pipeline_cmd
last_status=${PIPESTATUS[0]}
eval "cat ${_save_log_path}"
status_check $last_status "${pipeline_cmd}" "${status_log}"
sleep 2s
done
ps ux | grep -E 'web_service|pipeline' | awk '{print $2}' | xargs kill -s 9
done
done
else
echo "Does not support hardware other than CPU and GPU Currently!"
fi
done
fi
done
}
# set cuda device
GPUID=$2
if [ ${#GPUID} -le 0 ];then
env=" "
else
env="export CUDA_VISIBLE_DEVICES=${GPUID}"
fi
set CUDA_VISIBLE_DEVICES
eval $env
echo "################### run test ###################"
export Count=0
IFS="|"
func_serving "${web_service_cmd}"
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或