mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2025-10-06 00:57:33 +08:00
[Other] PPOCR models support model clone function (#1072)
* Refactor PaddleSeg with preprocessor && postprocessor * Fix bugs * Delete redundancy code * Modify by comments * Refactor according to comments * Add batch evaluation * Add single test script * Add ppliteseg single test script && fix eval(raise) error * fix bug * Fix evaluation segmentation.py batch predict * Fix segmentation evaluation bug * Fix evaluation segmentation bugs * Update segmentation result docs * Update old predict api and DisableNormalizeAndPermute * Update resize segmentation label map with cv::INTER_NEAREST * Add Model Clone function for PaddleClas && PaddleDet && PaddleSeg * Add multi thread demo * Add python model clone function * Add multi thread python && C++ example * Fix bug * Update python && cpp multi_thread examples * Add cpp && python directory * Add README.md for examples * Delete redundant code * Create README_CN.md * Rename README_CN.md to README.md * Update README.md * Update README.md * Update VERSION_NUMBER * Update requirements.txt * Update README.md * update version in doc: * [Serving]Update Dockerfile (#1037) Update Dockerfile * Add license notice for RVM onnx model file (#1060) * [Model] Add GPL-3.0 license (#1065) Add GPL-3.0 license * PPOCR model support model clone * Update README.md * Update PPOCRv2 && PPOCRv3 clone code * Update PPOCR python __init__ * Add multi thread ocr example code * Update README.md * Update README.md * Update ResNet50_vd_infer multi process code * Add PPOCR multi process && thread example * Update README.md * Update README.md * Update multi-thread docs Co-authored-by: Jason <jiangjiajun@baidu.com> Co-authored-by: leiqing <54695910+leiqing1@users.noreply.github.com> Co-authored-by: heliqi <1101791222@qq.com> Co-authored-by: WJJ1995 <wjjisloser@163.com>
This commit is contained in:
huangjianhui
14
tutorials/multi_thread/cpp/pipeline/CMakeLists.txt
Normal file
14
tutorials/multi_thread/cpp/pipeline/CMakeLists.txt
Normal file
@@ -0,0 +1,14 @@
|
||||
PROJECT(multi_thread_demo C CXX)
|
||||
CMAKE_MINIMUM_REQUIRED (VERSION 3.10)
|
||||
|
||||
# 指定下载解压后的fastdeploy库路径
|
||||
option(FASTDEPLOY_INSTALL_DIR "Path of downloaded fastdeploy sdk.")
|
||||
|
||||
include(${FASTDEPLOY_INSTALL_DIR}/FastDeploy.cmake)
|
||||
|
||||
# 添加FastDeploy依赖头文件
|
||||
include_directories(${FASTDEPLOY_INCS})
|
||||
|
||||
add_executable(multi_thread_demo ${PROJECT_SOURCE_DIR}/multi_thread_ocr.cc)
|
||||
# 添加FastDeploy库依赖
|
||||
target_link_libraries(multi_thread_demo ${FASTDEPLOY_LIBS} pthread)
|
||||
59
tutorials/multi_thread/cpp/pipeline/README.md
Normal file
59
tutorials/multi_thread/cpp/pipeline/README.md
Normal file
@@ -0,0 +1,59 @@
|
||||
English | [简体中文](README_CN.md)
|
||||
# PPOCRv3 C++ multi-thread Deployment Example
|
||||
|
||||
This directory provides examples file `multi_thread_ocr.cc` to fast deploy PPOCRv3 on CPU/GPU and GPU accelerated by TensorRT.
|
||||
|
||||
Two steps before deployment
|
||||
|
||||
- 1. Software and hardware should meet the requirements. Please refer to [FastDeploy Environment Requirements](../../../../docs/cn/build_and_install/download_prebuilt_libraries.md)
|
||||
- 2. Download the precompiled deployment library and samples code according to your development environment. Refer to [FastDeploy Precompiled Library](../../../../docs/cn/build_and_install/download_prebuilt_libraries.md)
|
||||
|
||||
Taking the PPOCRv3 inference on Linux as an example, the compilation test can be completed by executing the following command in this directory. FastDeploy version 0.7.0 or above (x.x.x>=0.7.0) is required to support this model.
|
||||
|
||||
```bash
|
||||
mkdir build
|
||||
cd build
|
||||
# Download the FastDeploy precompiled library. Users can choose your appropriate version in the `FastDeploy Precompiled Library` mentioned above
|
||||
wget https://bj.bcebos.com/fastdeploy/release/cpp/fastdeploy-linux-x64-x.x.x.tgz
|
||||
tar xvf fastdeploy-linux-x64-x.x.x.tgz
|
||||
cmake .. -DFASTDEPLOY_INSTALL_DIR=${PWD}/fastdeploy-linux-x64-x.x.x
|
||||
make -j
|
||||
|
||||
|
||||
# Download model, image, and dictionary files
|
||||
wget https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_infer.tar
|
||||
tar -xvf ch_PP-OCRv3_det_infer.tar
|
||||
|
||||
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar
|
||||
tar -xvf ch_ppocr_mobile_v2.0_cls_infer.tar
|
||||
|
||||
wget https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_infer.tar
|
||||
tar -xvf ch_PP-OCRv3_rec_infer.tar
|
||||
|
||||
wget https://gitee.com/paddlepaddle/PaddleOCR/raw/release/2.6/doc/imgs/12.jpg
|
||||
|
||||
wget https://gitee.com/paddlepaddle/PaddleOCR/raw/release/2.6/ppocr/utils/ppocr_keys_v1.txt
|
||||
|
||||
# CPU multi-thread inference
|
||||
./multi_thread_demo ./ch_PP-OCRv3_det_infer ./ch_ppocr_mobile_v2.0_cls_infer ./ch_PP-OCRv3_rec_infer ./ppocr_keys_v1.txt ./12.jpg 0 1
|
||||
# GPU multi-thread inference
|
||||
./multi_thread_demo ./ch_PP-OCRv3_det_infer ./ch_ppocr_mobile_v2.0_cls_infer ./ch_PP-OCRv3_rec_infer ./ppocr_keys_v1.txt ./12.jpg 1 1
|
||||
# TensorRT multi-thread inference on GPU
|
||||
./multi_thread_demo ./ch_PP-OCRv3_det_infer ./ch_ppocr_mobile_v2.0_cls_infer ./ch_PP-OCRv3_rec_infer ./ppocr_keys_v1.txt ./12.jpg 2 1
|
||||
# Paddle-TRT multi-thread inference on GPU
|
||||
./multi_thread_demo ./ch_PP-OCRv3_det_infer ./ch_ppocr_mobile_v2.0_cls_infer ./ch_PP-OCRv3_rec_infer ./ppocr_keys_v1.txt ./12.jpg 3 1
|
||||
# KunlunXin XPU multi-thread inference
|
||||
./multi_thread_demo ./ch_PP-OCRv3_det_infer ./ch_ppocr_mobile_v2.0_cls_infer ./ch_PP-OCRv3_rec_infer ./ppocr_keys_v1.txt ./12.jpg 4 1
|
||||
>> **Notice**: the last number in above command is thread number
|
||||
|
||||
The above command works for Linux or MacOS. For SDK in Windows, refer to:
|
||||
- [How to use FastDeploy C++ SDK in Windows](../../../docs/cn/faq/use_sdk_on_windows.md)
|
||||
|
||||
The result returned after running is as follows
|
||||
```
|
||||
Thread Id: 0
|
||||
det boxes: [[42,413],[483,391],[484,428],[43,450]]rec text: 上海斯格威铂尔大酒店 rec score:0.980085 cls label: 0 cls score: 1.000000
|
||||
det boxes: [[187,456],[399,448],[400,480],[188,488]]rec text: 打浦路15号 rec score:0.964993 cls label: 0 cls score: 1.000000
|
||||
det boxes: [[23,507],[513,488],[515,529],[24,548]]rec text: 绿洲仕格维花园公寓 rec score:0.993727 cls label: 0 cls score: 1.000000
|
||||
det boxes: [[74,553],[427,542],[428,571],[75,582]]rec text: 打浦路252935号 rec score:0.947723 cls label: 0 cls score: 1.000000
|
||||
```
|
||||
59
tutorials/multi_thread/cpp/pipeline/README_CN.md
Normal file
59
tutorials/multi_thread/cpp/pipeline/README_CN.md
Normal file
@@ -0,0 +1,59 @@
|
||||
[English](README.md) | 中文
|
||||
# PPOCRv3模型 C++多线程部署示例
|
||||
|
||||
本目录下提供`multi_thread_ocr.cc`快速完成PPOCRv3系列模型在CPU/GPU，以及GPU上通过TensorRT加速多线程部署的示例。
|
||||
|
||||
在部署前，需确认以下两个步骤
|
||||
|
||||
- 1. 软硬件环境满足要求，参考[FastDeploy环境要求](../../../../docs/cn/build_and_install/download_prebuilt_libraries.md)
|
||||
- 2. 根据开发环境，下载预编译部署库和samples代码，参考[FastDeploy预编译库](../../../../docs/cn/build_and_install/download_prebuilt_libraries.md)
|
||||
|
||||
以Linux上ResNet50_vd推理为例，在本目录执行如下命令即可完成编译测试，支持此模型需保证FastDeploy版本0.7.0以上(x.x.x>=0.7.0)
|
||||
|
||||
```bash
|
||||
mkdir build
|
||||
cd build
|
||||
# 下载FastDeploy预编译库，用户可在上文提到的`FastDeploy预编译库`中自行选择合适的版本使用
|
||||
wget https://bj.bcebos.com/fastdeploy/release/cpp/fastdeploy-linux-x64-x.x.x.tgz
|
||||
tar xvf fastdeploy-linux-x64-x.x.x.tgz
|
||||
cmake .. -DFASTDEPLOY_INSTALL_DIR=${PWD}/fastdeploy-linux-x64-x.x.x
|
||||
make -j
|
||||
|
||||
|
||||
# 下载模型,图片和字典文件
|
||||
wget https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_infer.tar
|
||||
tar -xvf ch_PP-OCRv3_det_infer.tar
|
||||
|
||||
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar
|
||||
tar -xvf ch_ppocr_mobile_v2.0_cls_infer.tar
|
||||
|
||||
wget https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_infer.tar
|
||||
tar -xvf ch_PP-OCRv3_rec_infer.tar
|
||||
|
||||
wget https://gitee.com/paddlepaddle/PaddleOCR/raw/release/2.6/doc/imgs/12.jpg
|
||||
|
||||
wget https://gitee.com/paddlepaddle/PaddleOCR/raw/release/2.6/ppocr/utils/ppocr_keys_v1.txt
|
||||
|
||||
# CPU推理
|
||||
./multi_thread_demo ./ch_PP-OCRv3_det_infer ./ch_ppocr_mobile_v2.0_cls_infer ./ch_PP-OCRv3_rec_infer ./ppocr_keys_v1.txt ./12.jpg 0 1
|
||||
# GPU推理
|
||||
./multi_thread_demo ./ch_PP-OCRv3_det_infer ./ch_ppocr_mobile_v2.0_cls_infer ./ch_PP-OCRv3_rec_infer ./ppocr_keys_v1.txt ./12.jpg 1 1
|
||||
# GPU上TensorRT推理
|
||||
./multi_thread_demo ./ch_PP-OCRv3_det_infer ./ch_ppocr_mobile_v2.0_cls_infer ./ch_PP-OCRv3_rec_infer ./ppocr_keys_v1.txt ./12.jpg 2 1
|
||||
# GPU上Paddle-TRT推理
|
||||
./multi_thread_demo ./ch_PP-OCRv3_det_infer ./ch_ppocr_mobile_v2.0_cls_infer ./ch_PP-OCRv3_rec_infer ./ppocr_keys_v1.txt ./12.jpg 3 1
|
||||
# 昆仑芯XPU推理
|
||||
./multi_thread_demo ./ch_PP-OCRv3_det_infer ./ch_ppocr_mobile_v2.0_cls_infer ./ch_PP-OCRv3_rec_infer ./ppocr_keys_v1.txt ./12.jpg 4 1
|
||||
>> **注意**: 最后一位数字表示线程数
|
||||
|
||||
以上命令只适用于Linux或MacOS, Windows下SDK的使用方式请参考:
|
||||
- [如何在Windows中使用FastDeploy C++ SDK](../../../docs/cn/faq/use_sdk_on_windows.md)
|
||||
|
||||
运行完成后返回结果如下所示
|
||||
```
|
||||
Thread Id: 0
|
||||
det boxes: [[42,413],[483,391],[484,428],[43,450]]rec text: 上海斯格威铂尔大酒店 rec score:0.980085 cls label: 0 cls score: 1.000000
|
||||
det boxes: [[187,456],[399,448],[400,480],[188,488]]rec text: 打浦路15号 rec score:0.964993 cls label: 0 cls score: 1.000000
|
||||
det boxes: [[23,507],[513,488],[515,529],[24,548]]rec text: 绿洲仕格维花园公寓 rec score:0.993727 cls label: 0 cls score: 1.000000
|
||||
det boxes: [[74,553],[427,542],[428,571],[75,582]]rec text: 打浦路252935号 rec score:0.947723 cls label: 0 cls score: 1.000000
|
||||
```
|
||||
177
tutorials/multi_thread/cpp/pipeline/multi_thread_ocr.cc
Executable file
177
tutorials/multi_thread/cpp/pipeline/multi_thread_ocr.cc
Executable file
@@ -0,0 +1,177 @@
|
||||
// Copyright (c) 2022 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.
|
||||
|
||||
#include <thread>
|
||||
#include "fastdeploy/vision.h"
|
||||
#ifdef WIN32
|
||||
const char sep = '\\';
|
||||
#else
|
||||
const char sep = '/';
|
||||
#endif
|
||||
|
||||
void Predict(fastdeploy::pipeline::PPOCRv3 *model, int thread_id, const std::vector<std::string>& images) {
|
||||
for (auto const &image_file : images) {
|
||||
auto im = cv::imread(image_file);
|
||||
|
||||
fastdeploy::vision::OCRResult res;
|
||||
if (!model->Predict(im, &res)) {
|
||||
std::cerr << "Failed to predict." << std::endl;
|
||||
return;
|
||||
}
|
||||
|
||||
// print res
|
||||
std::cout << "Thread Id: " << thread_id << std::endl;
|
||||
std::cout << res.Str() << std::endl;
|
||||
}
|
||||
}
|
||||
|
||||
void GetImageList(std::vector<std::vector<std::string>>* image_list, const std::string& image_file_path, int thread_num){
|
||||
std::vector<cv::String> images;
|
||||
cv::glob(image_file_path, images, false);
|
||||
// number of image files in images folder
|
||||
size_t count = images.size();
|
||||
size_t num = count / thread_num;
|
||||
for (int i = 0; i < thread_num; i++) {
|
||||
std::vector<std::string> temp_list;
|
||||
if (i == thread_num - 1) {
|
||||
for (size_t j = i*num; j < count; j++){
|
||||
temp_list.push_back(images[j]);
|
||||
}
|
||||
} else {
|
||||
for (size_t j = 0; j < num; j++){
|
||||
temp_list.push_back(images[i * num + j]);
|
||||
}
|
||||
}
|
||||
(*image_list)[i] = temp_list;
|
||||
}
|
||||
}
|
||||
|
||||
void InitAndInfer(const std::string& det_model_dir, const std::string& cls_model_dir, const std::string& rec_model_dir, const std::string& rec_label_file, const std::string& image_file_path, const fastdeploy::RuntimeOption& option, int thread_num) {
|
||||
auto det_model_file = det_model_dir + sep + "inference.pdmodel";
|
||||
auto det_params_file = det_model_dir + sep + "inference.pdiparams";
|
||||
|
||||
auto cls_model_file = cls_model_dir + sep + "inference.pdmodel";
|
||||
auto cls_params_file = cls_model_dir + sep + "inference.pdiparams";
|
||||
|
||||
auto rec_model_file = rec_model_dir + sep + "inference.pdmodel";
|
||||
auto rec_params_file = rec_model_dir + sep + "inference.pdiparams";
|
||||
|
||||
auto det_option = option;
|
||||
auto cls_option = option;
|
||||
auto rec_option = option;
|
||||
|
||||
// The cls and rec model can inference a batch of images now.
|
||||
// User could initialize the inference batch size and set them after create PP-OCR model.
|
||||
int cls_batch_size = 1;
|
||||
int rec_batch_size = 6;
|
||||
|
||||
// If use TRT backend, the dynamic shape will be set as follow.
|
||||
// We recommend that users set the length and height of the detection model to a multiple of 32.
|
||||
// We also recommend that users set the Trt input shape as follow.
|
||||
det_option.SetTrtInputShape("x", {1, 3, 64,64}, {1, 3, 640, 640},
|
||||
{1, 3, 960, 960});
|
||||
cls_option.SetTrtInputShape("x", {1, 3, 48, 10}, {cls_batch_size, 3, 48, 320}, {cls_batch_size, 3, 48, 1024});
|
||||
rec_option.SetTrtInputShape("x", {1, 3, 48, 10}, {rec_batch_size, 3, 48, 320},
|
||||
{rec_batch_size, 3, 48, 2304});
|
||||
|
||||
// Users could save TRT cache file to disk as follow.
|
||||
// det_option.SetTrtCacheFile(det_model_dir + sep + "det_trt_cache.trt");
|
||||
// cls_option.SetTrtCacheFile(cls_model_dir + sep + "cls_trt_cache.trt");
|
||||
// rec_option.SetTrtCacheFile(rec_model_dir + sep + "rec_trt_cache.trt");
|
||||
|
||||
auto det_model = fastdeploy::vision::ocr::DBDetector(det_model_file, det_params_file, det_option);
|
||||
auto cls_model = fastdeploy::vision::ocr::Classifier(cls_model_file, cls_params_file, cls_option);
|
||||
auto rec_model = fastdeploy::vision::ocr::Recognizer(rec_model_file, rec_params_file, rec_label_file, rec_option);
|
||||
|
||||
assert(det_model.Initialized());
|
||||
assert(cls_model.Initialized());
|
||||
assert(rec_model.Initialized());
|
||||
|
||||
// The classification model is optional, so the PP-OCR can also be connected in series as follows
|
||||
// auto ppocr_v3 = fastdeploy::pipeline::PPOCRv3(&det_model, &rec_model);
|
||||
auto ppocr_v3 = fastdeploy::pipeline::PPOCRv3(&det_model, &cls_model, &rec_model);
|
||||
|
||||
// Set inference batch size for cls model and rec model, the value could be -1 and 1 to positive infinity.
|
||||
// When inference batch size is set to -1, it means that the inference batch size
|
||||
// of the cls and rec models will be the same as the number of boxes detected by the det model.
|
||||
ppocr_v3.SetClsBatchSize(cls_batch_size);
|
||||
ppocr_v3.SetRecBatchSize(rec_batch_size);
|
||||
|
||||
if(!ppocr_v3.Initialized()){
|
||||
std::cerr << "Failed to initialize PP-OCR." << std::endl;
|
||||
return;
|
||||
}
|
||||
|
||||
|
||||
std::vector<decltype(ppocr_v3.Clone())> models;
|
||||
for (int i = 0; i < thread_num; ++i) {
|
||||
models.emplace_back(std::move(ppocr_v3.Clone()));
|
||||
}
|
||||
|
||||
std::vector<std::vector<std::string>> image_list(thread_num);
|
||||
GetImageList(&image_list, image_file_path, thread_num);
|
||||
|
||||
std::vector<std::thread> threads;
|
||||
for (int i = 0; i < thread_num; ++i) {
|
||||
threads.emplace_back(Predict, models[i].get(), i, image_list[i]);
|
||||
}
|
||||
|
||||
for (int i = 0; i < thread_num; ++i) {
|
||||
threads[i].join();
|
||||
}
|
||||
}
|
||||
|
||||
int main(int argc, char* argv[]) {
|
||||
if (argc < 7) {
|
||||
std::cout << "Usage: infer_demo path/to/det_model path/to/cls_model "
|
||||
"path/to/rec_model path/to/rec_label_file path/to/image "
|
||||
"run_option thread_num,"
|
||||
"e.g ./infer_demo ./ch_PP-OCRv3_det_infer "
|
||||
"./ch_ppocr_mobile_v2.0_cls_infer ./ch_PP-OCRv3_rec_infer "
|
||||
"./ppocr_keys_v1.txt ./12.jpg 0 3"
|
||||
<< std::endl;
|
||||
std::cout << "The data type of run_option is int, 0: run with cpu; 1: run "
|
||||
"with gpu; 2: run with gpu and use tensorrt backend; 3: run with gpu and use Paddle-TRT; 4: run with kunlunxin."
|
||||
<< std::endl;
|
||||
return -1;
|
||||
}
|
||||
|
||||
fastdeploy::RuntimeOption option;
|
||||
int flag = std::atoi(argv[6]);
|
||||
|
||||
if (flag == 0) {
|
||||
option.UseCpu();
|
||||
} else if (flag == 1) {
|
||||
option.UseGpu();
|
||||
} else if (flag == 2) {
|
||||
option.UseGpu();
|
||||
option.UseTrtBackend();
|
||||
} else if (flag == 3) {
|
||||
option.UseGpu();
|
||||
option.UseTrtBackend();
|
||||
option.EnablePaddleTrtCollectShape();
|
||||
option.EnablePaddleToTrt();
|
||||
} else if (flag == 4) {
|
||||
option.UseKunlunXin();
|
||||
}
|
||||
|
||||
std::string det_model_dir = argv[1];
|
||||
std::string cls_model_dir = argv[2];
|
||||
std::string rec_model_dir = argv[3];
|
||||
std::string rec_label_file = argv[4];
|
||||
std::string image_file_path = argv[5];
|
||||
int thread_num = std::atoi(argv[7]);
|
||||
InitAndInfer(det_model_dir, cls_model_dir, rec_model_dir, rec_label_file, image_file_path, option, thread_num);
|
||||
return 0;
|
||||
}
|
||||
Reference in New Issue
Block a user