Skip to content

GitLab

"sgl-router/src/git@developer.sourcefind.cn:change/sglang.git" did not exist on "1f76fc874759c257b4e928b9847a8da8e8ea2b30"
Commit 799d0dc5 authored by LDOUBLEV's avatar LDOUBLEV
Browse files

Merge branch 'dygraph' of https://github.com/PaddlePaddle/PaddleOCR into test_v11

parents 1d7c9f81 c81baca7
Hide whitespace changes
Inline Side-by-side
Showing with 32 additions and 28 deletions
doc/doc_ch/inference.md
View file @ 799d0dc5
......@@ -34,7 +34,7 @@ inference 模型（`paddle.jit.save`保存的模型）
- [1. 超轻量中文OCR模型推理](#超轻量中文OCR模型推理)
- [2. 其他模型推理](#其他模型推理)
- [六、参数解释](参数解释)
- [六、参数解释](#参数解释)
<a name="训练模型转inference模型"></a>
......@@ -504,7 +504,7 @@ PSE算法相关参数如下
| e2e_model_dir | str | 无，如果使用端到端模型，该项是必填项 | 端到端模型inference模型路径 |
| e2e_limit_side_len | int | 768 | 端到端的输入图像边长限制 |
| e2e_limit_type | str | "max" | 端到端的边长限制类型，目前支持`min`, `max`，`min`表示保证图像最短边不小于`e2e_limit_side_len`，`max`表示保证图像最长边不大于`e2e_limit_side_len` |
| e2e_pgnet_score_thresh | float | xx | xx |
| e2e_pgnet_score_thresh | float | 0.5 | 端到端得分阈值，小于该阈值的结果会被丢弃 |
| e2e_char_dict_path | str | "./ppocr/utils/ic15_dict.txt" | 识别的字典文件路径 |
| e2e_pgnet_valid_set | str | "totaltext" | 验证集名称，目前支持`totaltext`, `partvgg`，不同数据集对应的后处理方式不同，与训练过程保持一致即可 |
| e2e_pgnet_mode | str | "fast" | PGNet的检测结果得分计算方法，支持`fast`和`slow`，`fast`是根据polygon的外接矩形边框内的所有像素计算平均得分，`slow`是根据原始polygon内的所有像素计算平均得分，计算速度相对较慢一些，但是更加准确一些。 |
......
doc/doc_ch/pgnet.md
View file @ 799d0dc5
......@@ -66,13 +66,13 @@ wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/pgnet/e2e_server_pgnetA_infer.
### 单张图像或者图像集合预测
```bash
# 预测image_dir指定的单张图像
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e_server_pgnetA_infer/" --e2e_pgnet_polygon=True
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e_server_pgnetA_infer/" --e2e_pgnet_valid_set="totaltext"
# 预测image_dir指定的图像集合
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/" --e2e_model_dir="./inference/e2e_server_pgnetA_infer/" --e2e_pgnet_polygon=True
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/" --e2e_model_dir="./inference/e2e_server_pgnetA_infer/" --e2e_pgnet_valid_set="totaltext"
# 如果想使用CPU进行预测，需设置use_gpu参数为False
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e_server_pgnetA_infer/" --e2e_pgnet_polygon=True --use_gpu=False
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e_server_pgnetA_infer/" --e2e_pgnet_valid_set="totaltext" --use_gpu=False
```
### 可视化结果
可视化文本检测结果默认保存到./inference_results文件夹里面，结果文件的名称前缀为'e2e_res'。结果示例如下：
......@@ -167,9 +167,9 @@ python3 tools/infer_e2e.py -c configs/e2e/e2e_r50_vd_pg.yml -o Global.infer_img=
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/pgnet/en_server_pgnetA.tar && tar xf en_server_pgnetA.tar
python3 tools/export_model.py -c configs/e2e/e2e_r50_vd_pg.yml -o Global.pretrained_model=./en_server_pgnetA/best_accuracy Global.load_static_weights=False Global.save_inference_dir=./inference/e2e
```
**PGNet端到端模型推理，需要设置参数`--e2e_algorithm="PGNet"`**，可以执行如下命令：
**PGNet端到端模型推理，需要设置参数`--e2e_algorithm="PGNet"` and `--e2e_pgnet_valid_set="partvgg"`**，可以执行如下命令：
```
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img_10.jpg" --e2e_model_dir="./inference/e2e/" --e2e_pgnet_polygon=False
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img_10.jpg" --e2e_model_dir="./inference/e2e/" --e2e_pgnet_valid_set="partvgg" --e2e_pgnet_valid_set="totaltext"
```
可视化文本检测结果默认保存到`./inference_results`文件夹里面，结果文件的名称前缀为'e2e_res'。结果示例如下：
......@@ -178,9 +178,9 @@ python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/im
#### (2). 弯曲文本检测模型（Total-Text）
对于弯曲文本样例
**PGNet端到端模型推理，需要设置参数`--e2e_algorithm="PGNet"`，同时，还需要增加参数`--e2e_pgnet_polygon=True`，**可以执行如下命令：
**PGNet端到端模型推理，需要设置参数`--e2e_algorithm="PGNet"`，同时，还需要增加参数`--e2e_pgnet_valid_set="totaltext"`，**可以执行如下命令：
```
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e/" --e2e_pgnet_polygon=True
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e/" --e2e_pgnet_valid_set="totaltext"
```
可视化文本端到端结果默认保存到`./inference_results`文件夹里面，结果文件的名称前缀为'e2e_res'。结果示例如下：
......
doc/doc_en/pgnet_en.md
View file @ 799d0dc5
......@@ -59,13 +59,13 @@ After decompression, there should be the following file structure:
### Single image or image set prediction
```bash
# Prediction single image specified by image_dir
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e_server_pgnetA_infer/" --e2e_pgnet_polygon=True
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e_server_pgnetA_infer/" --e2e_pgnet_valid_set="totaltext"
# Prediction the collection of images specified by image_dir
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/" --e2e_model_dir="./inference/e2e_server_pgnetA_infer/" --e2e_pgnet_polygon=True
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/" --e2e_model_dir="./inference/e2e_server_pgnetA_infer/" --e2e_pgnet_valid_set="totaltext"
# If you want to use CPU for prediction, you need to set use_gpu parameter is false
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e_server_pgnetA_infer/" --e2e_pgnet_polygon=True --use_gpu=False
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e_server_pgnetA_infer/" --use_gpu=False --e2e_pgnet_valid_set="totaltext"
```
### Visualization results
The visualized end-to-end results are saved to the `./inference_results` folder by default, and the name of the result file is prefixed with 'e2e_res'. Examples of results are as follows:
......@@ -166,9 +166,9 @@ First, convert the model saved in the PGNet end-to-end training process into an
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/pgnet/en_server_pgnetA.tar && tar xf en_server_pgnetA.tar
python3 tools/export_model.py -c configs/e2e/e2e_r50_vd_pg.yml -o Global.pretrained_model=./en_server_pgnetA/best_accuracy Global.load_static_weights=False Global.save_inference_dir=./inference/e2e
```
**For PGNet quadrangle end-to-end model inference, you need to set the parameter `--e2e_algorithm="PGNet"`**, run the following command:
**For PGNet quadrangle end-to-end model inference, you need to set the parameter `--e2e_algorithm="PGNet"` and `--e2e_pgnet_valid_set="partvgg"`**, run the following command:
```
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img_10.jpg" --e2e_model_dir="./inference/e2e/" --e2e_pgnet_polygon=False
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img_10.jpg" --e2e_model_dir="./inference/e2e/" --e2e_pgnet_valid_set="partvgg"
```
The visualized text detection results are saved to the `./inference_results` folder by default, and the name of the result file is prefixed with 'e2e_res'. Examples of results are as follows:
......@@ -176,9 +176,9 @@ The visualized text detection results are saved to the `./inference_results` fol
#### (2). Curved text detection model (Total-Text)
For the curved text example, we use the same model as the quadrilateral
**For PGNet end-to-end curved text detection model inference, you need to set the parameter `--e2e_algorithm="PGNet"` and `--e2e_pgnet_polygon=True`**, run the following command:
**For PGNet end-to-end curved text detection model inference, you need to set the parameter `--e2e_algorithm="PGNet"` and `--e2e_pgnet_valid_set="totaltext"`**, run the following command:
```
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e/" --e2e_pgnet_polygon=True
python3 tools/infer/predict_e2e.py --e2e_algorithm="PGNet" --image_dir="./doc/imgs_en/img623.jpg" --e2e_model_dir="./inference/e2e/" --e2e_pgnet_valid_set="totaltext"
```
The visualized text detection results are saved to the `./inference_results` folder by default, and the name of the result file is prefixed with 'e2e_res'. Examples of results are as follows:
......
test_tipc/configs/ch_ppocr_mobile_v2.0_det/model_linux_gpu_normal_normal_serving_python_linux_gpu_cpu.txt
View file @ 799d0dc5
......@@ -15,4 +15,4 @@ op.det.local_service_conf.thread_num:1|6
op.det.local_service_conf.use_trt:False|True
op.det.local_service_conf.precision:fp32|fp16|int8
pipline:pipeline_rpc_client.py|pipeline_http_client.py
--image_dir:../../doc/imgs
\ No newline at end of file
--image_dir:../../doc/imgs
test_tipc/prepare.sh
View file @ 799d0dc5
......@@ -201,8 +201,11 @@ fi
if [ ${MODE} = "serving_infer" ];then
# prepare serving env
python_name=$(func_parser_value "${lines[2]}")
wget https://paddle-serving.bj.bcebos.com/chain/paddle_serving_server_gpu-0.0.0.post101-py3-none-any.whl
python_name_list=$(func_parser_value "${lines[2]}")
IFS='|'
array=(${python_name_list})
python_name=${array[0]}
wget -nc https://paddle-serving.bj.bcebos.com/chain/paddle_serving_server_gpu-0.0.0.post101-py3-none-any.whl
${python_name} -m pip install install paddle_serving_server_gpu-0.0.0.post101-py3-none-any.whl
${python_name} -m pip install paddle_serving_client==0.6.1
${python_name} -m pip install paddle-serving-app==0.6.3
......
test_tipc/readme.md
View file @ 799d0dc5
# 飞桨训推一体认证（TIPC）
# 飞桨训推一体全流程（TIPC）
## 1. 简介
飞桨除了基本的模型训练和预测，还提供了支持多端多平台的高性能推理部署工具。本文档提供了PaddleOCR中所有模型的飞桨训推一体认证 (Training and Inference Pipeline Certification(TIPC)) 信息和测试工具，方便用户查阅每种模型的训练推理部署打通情况，并可以进行一键测试。
飞桨除了基本的模型训练和预测，还提供了支持多端多平台的高性能推理部署工具。本文档提供了PaddleOCR中所有模型的飞桨训推一体全流程（Training and Inference Pipeline Criterion(TIPC)）信息和测试工具，方便用户查阅每种模型的训练推理部署打通情况，并可以进行一键测试。
<div align="center">
<img src="docs/guide.png" width="1000">
......
test_tipc/test_serving.sh
View file @ 799d0dc5
......@@ -10,7 +10,7 @@ lines=(${dataline})
# parser serving
model_name=$(func_parser_value "${lines[1]}")
python=$(func_parser_value "${lines[2]}")
python_list=$(func_parser_value "${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]}")
......@@ -54,14 +54,15 @@ function func_serving(){
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} ${set_serving_client}"
python_list=(${python_list})
trans_model_cmd="${python_list[0]} ${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 python in ${python_list[*]}; 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"
......@@ -91,9 +92,6 @@ function func_serving(){
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} ${set_cpu_threads} &"
......@@ -124,6 +122,9 @@ function func_serving(){
continue
fi
set_tensorrt=$(func_set_params "${web_use_trt_key}" "${use_trt}")
if [ ${use_trt} = True ]; then
device_type=2
fi
set_precision=$(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
......
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