PaddleOCR

https://github.com/ymylive/chaoxing-fanya
Admin
PaddleOCR is an industry-leading OCR and document AI engine that converts documents and images into...
Tokens:653,558
Snippets:7,366
Trust Score:7.2
Update:4 months ago
Show doc for...
Tokens:
Raw
### Example Deployment Steps

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/deployment/on_device_deployment.md

A consolidated example of the deployment steps.

```shell
# 1. Download necessary Paddle Lite prediction library
cd PaddleX-Lite-Deploy/libs
sh download.sh

# 2. Download paddle_lite_opt tool optimized nb model files and prediction images, dictionary files, etc.
cd ../ocr/assets
sh download.sh PP-OCRv5_mobile

# 3. Complete the compilation of the executable file
cd ../android/shell/ppocr_demo
sh build.sh

# 4. Prediction
sh run.sh PP-OCRv5_mobile
```

--------------------------------

### Install PaddlePaddle (CPU)

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/paddlex/quick_start.en.md

Command to install the CPU version of PaddlePaddle.

```bash
python -m pip install paddlepaddle==3.0.0 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
```

--------------------------------

### Install PaddleX

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/paddlex/quick_start.en.md

Command to install PaddleX with OCR support.

```bash
pip install "paddlex[ocr]"
```

--------------------------------

### PP-StructureV3 Command Line Usage

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/quick_start.en.md

Example command for PP-StructureV3.

```bash
paddleocr pp_structurev3 -i ./pp_structure_v3_demo.png  --use_doc_orientation_classify False --use_doc_unwarping False
```

--------------------------------

### Manual Start - Backend

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/WEB_FRONTEND_GUIDE.md

Commands to install Python dependencies and start the backend service.

```bash
# 安装Python依赖（首次运行）
pip install -r requirements.txt

# 启动后端服务
python app.py
```

--------------------------------

### Manual Start - Frontend

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/WEB_FRONTEND_GUIDE.md

Commands to navigate to the frontend directory, install Node.js dependencies, and start the development server.

```bash
# 进入前端目录
cd web

# 安装Node.js依赖（首次运行）
npm install

# 启动开发服务器
npm run dev
```

--------------------------------

### TensorRT Installation Example

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/test_tipc/docs/install.md

Commands to set the LD_LIBRARY_PATH environment variable and install TensorRT and GraphSurgeon using pip.

```bash
# The following installation command uses '${version}' for the downloaded TensorRT version, e.g., 7.1.3.4
# Set environment variables, <TensorRT-${version}/lib> is the lib directory after unzipping TensorRT
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:<TensorRT-${version}/lib>

# Install TensorRT
cd TensorRT-${version}/python
pip3.7 install tensorrt-*-cp3x-none-linux_x86_64.whl

# Install graphsurgeon
cd TensorRT-${version}/graphsurgeon
```

--------------------------------

### Command-line Quick Start

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/pipeline_usage/PP-StructureV3.md

This section demonstrates how to quickly experience the PP-StructureV3 pipeline using a single command. It includes examples for specifying whether to use document orientation classification, text image unwarping, and text line orientation models.

```bash
paddleocr pp_structurev3 -i https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/pp_structure_v3_demo.png
```

```bash
paddleocr pp_structurev3 -i ./pp_structure_v3_demo.png --use_doc_orientation_classify True
```

```bash
paddleocr pp_structurev3 -i ./pp_structure_v3_demo.png --use_doc_unwarping True
```

```bash
paddleocr pp_structurev3 -i ./pp_structure_v3_demo.png --use_textline_orientation False
```

--------------------------------

### CUDNN Installation Example

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/test_tipc/docs/install.md

Steps to install CUDNN 8.1 using deb packages, including copying samples, compiling, and running a verification program. It also includes commands to install freeimage if needed.

```bash
# x.x.x represents the downloaded version number
# $HOME is the working directory
sudo dpkg -i libcudnn8_x.x.x-1+cudax.x_arm64.deb
sudo dpkg -i libcudnn8-dev_8.x.x.x-1+cudax.x_arm64.deb
sudo dpkg -i libcudnn8-samples_8.x.x.x-1+cudax.x_arm64.deb

# Verify correct installation
cp -r /usr/src/cudnn_samples_v8/ $HOME
cd  $HOME/cudnn_samples_v8/mnistCUDNN

# Compile
make clean && make
./mnistCUDNN

sudo apt-get install libfreeimage-dev
sudo apt-get install libfreeimage
```

--------------------------------

### Manual TensorRT Installation Example

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/deployment/high_performance_inference.en.md

Example commands for downloading, extracting, and installing TensorRT, and setting the library path.

```bash
# Download the TensorRT tar file
wget https://developer.nvidia.com/downloads/compute/machine-learning/tensorrt/secure/8.6.1/tars/TensorRT-8.6.1.6.Linux.x86_64-gnu.cuda-11.8.tar.gz
# Extract the TensorRT tar file
tar xvf TensorRT-8.6.1.6.Linux.x86_64-gnu.cuda-11.8.tar.gz
# Install the TensorRT wheel package
python -m pip install TensorRT-8.6.1.6/python/tensorrt-8.6.1-cp310-none-linux_x86_64.whl
# Add the absolute path of the TensorRT `lib` directory to LD_LIBRARY_PATH
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:TensorRT-8.6.1.6/lib"
```

--------------------------------

### Quick Start Command Line

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/module_usage/doc_vlm.en.md

A single command to quickly experience the DocVLM functionality with a sample image and query.

```bash
paddleocr doc_vlm -i "{'image': 'https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/medal_table.png', 'query': '识别这份表格的内容, 以markdown格式输出'}"
```

--------------------------------

### Install AutoLog

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/test_tipc/docs/install.md

Commands to clone the AutoLog repository, install its requirements, build a wheel, and then install the wheel.

```bash
# Install AutoLog
git clone https://github.com/LDOUBLEV/AutoLog
cd AutoLog
pip3.7 install -r requirements.txt
python3.7 setup.py bdist_wheel
pip3.7 install ./dist/auto_log-1.0.0-py3-none-any.whl
```

--------------------------------

### General CLI Usage

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/paddlex/quick_start.en.md

The unified CLI format for running pipelines in PaddleX.

```bash
paddlex --pipeline [Pipeline Name] --input [Input Image] --device [Running Device]
```

--------------------------------

### Install PaddleOCR Dependencies

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/test_tipc/docs/install.md

Commands to navigate into the PaddleOCR directory and install its requirements.

```bash
cd PaddleOCR
pip3.7 install -r requirements.txt
```

--------------------------------

### Manual Start Instructions

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/IMPLEMENTATION_SUMMARY.md

Instructions for manually starting the backend and frontend services in separate terminals.

```bash
# 后端
pip install -r requirements.txt
python app.py

# 前端（新终端）
cd web
npm install
npm run dev
```

--------------------------------

### OCR Pipeline Result Example

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/paddlex/quick_start.en.md

Example output from the OCR pipeline.

```json
{"res": {"input_path": "general_ocr_002.png", "page_index": null, "model_settings": {"use_doc_preprocessor": false, "use_textline_orientation": false}, "doc_preprocessor_res": {"input_path": null, "model_settings": {"use_doc_orientation_classify": true, "use_doc_unwarping": false}, "angle": 0},"dt_polys": [array([[ 3, 10],
       [82, 10],
       [82, 33],
       [ 3, 33]], dtype=int16), ...], "text_det_params": {"limit_side_len": 960, "limit_type": "max", "thresh": 0.3, "box_thresh": 0.6, "unclip_ratio": 2.0}, "text_type": "general", "textline_orientation_angles": [-1, ...], "text_rec_score_thresh": 0.0, "rec_texts": ["www.99*", ...], "rec_scores": [0.8980069160461426,  ...], "rec_polys": [array([[ 3, 10],
       [82, 10],
       [82, 33],
       [ 3, 33]], dtype=int16), ...], "rec_boxes": array([[  3,  10,  82,  33], ...], dtype=int16)}}
}
```

--------------------------------

### Quick Start (Recommended One-Click)

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/README.md

Clone the repository, navigate to the directory, and run the start script for Windows.

```bash
git clone --depth=1 https://github.com/ymylive/chaoxing-fanya
cd chaoxing-fanya
start.bat  # Windows 双击或命令行运行
```

--------------------------------

### Install PaddlePaddle (CPU)

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/quick_start.en.md

Command to install PaddlePaddle for CPU usage.

```bash
python -m pip install paddlepaddle==3.2.0 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
```

--------------------------------

### Quick Start Command Line

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/module_usage/text_image_unwarping.en.md

This command allows for a quick experience of the text image unwarping feature using a provided URL.

```bash
paddleocr text_image_unwarping -i https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/doc_test.jpg
```

--------------------------------

### Install PaddlePaddle

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/deploy/slim/auto_compression/README.md

Command to install PaddlePaddle CPU version.

```shell
python -m pip install paddlepaddle==2.5.1 -i https://pypi.tuna.tsinghua.edu.cn/simple
```

--------------------------------

### PP-StructureV3 Initialization

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/pipeline_usage/PP-StructureV3.md

Example of initializing the PP-StructureV3 pipeline, specifying the device for inference.

```python
pipeline = PPStructureV3(device="gpu") # 通过 device 指定模型推理时使用 GPU
```

--------------------------------

### Command-line Quick Start

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/pipeline_usage/doc_preprocessor.md

This snippet shows how to quickly experience the doc_preprocessor pipeline using the command line. It includes examples for basic usage, enabling document orientation classification, and enabling text image unwarping.

```bash
paddleocr doc_preprocessor -i https://paddle-model-ecology.bj.bcebos.com/paddlex/demo_image/doc_test_rotated.jpg

# 通过 --use_doc_orientation_classify 指定是否使用文档方向分类模型
paddleocr doc_preprocessor -i ./doc_test_rotated.jpg --use_doc_orientation_classify True

# 通过 --use_doc_unwarping 指定是否使用文本图像矫正模块
paddleocr doc_preprocessor -i ./doc_test_rotated.jpg --use_doc_unwarping True

```

--------------------------------

### Install PaddlePaddle (GPU)

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/paddlex/quick_start.en.md

Command to install the GPU version of PaddlePaddle, specifying CUDA 11.8.

```bash
python -m pip install paddlepaddle-gpu==3.0.0 -i https://www.paddlepaddle.org.cn/packages/stable/cu118/
```

--------------------------------

### Quick Start Command Line

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/module_usage/table_cells_detection.md

A single command to quickly experience table cells detection using a provided URL.

```bash
paddleocr table_cells_detection -i https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/table_recognition.jpg
```

--------------------------------

### Install PaddlePaddle (GPU)

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/quick_start.en.md

Command to install PaddlePaddle for GPU usage on Linux with CUDA 11.8.

```bash
python -m pip install paddlepaddle-gpu==3.2.0 -i https://www.paddlepaddle.org.cn/packages/stable/cu118/
```

--------------------------------

### Quick Start Command Line

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/module_usage/table_classification.md

A single command to quickly experience table classification.

```bash
paddleocr table_classification -i https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/table_recognition.jpg
```

--------------------------------

### Docker Image Installation

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/test_tipc/docs/install.md

Command to create a Docker image and map the current directory to /paddle within the image, followed by installing Paddle with TensorRT.

```bash
nvidia-docker run --name paddle -it -v $PWD:/paddle paddlepaddle/paddle:latest-dev-cuda10.1-cudnn7-gcc82 /bin/bash
cd /paddle

# Install paddle with TRT
pip3.7 install https://paddle-wheel.bj.bcebos.com/with-trt/2.1.3/linux-gpu-cuda10.1-cudnn7-mkl-gcc8.2-trt6-avx/paddlepaddle_gpu-2.1.3.post101-cp37-cp37m-linux_x86_64.whl
```

--------------------------------

### Install Service Module Examples

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/deploy/hubserving/readme_en.md

Commands to install various PaddleOCR service modules using `hub install` on Linux. Replace `/` with `\` for Windows.

```bash
hub install deploy/hubserving/ocr_det
hub install deploy/hubserving/ocr_cls
hub install deploy/hubserving/ocr_rec
hub install deploy/hubserving/ocr_system
hub install deploy/hubserving/structure_table
hub install deploy/hubserving/structure_system
hub install deploy/hubserving/kie_ser
hub install deploy/hubserving/kie_ser_re
```

--------------------------------

### Quick Start Command Line

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/module_usage/text_detection.md

A single command to quickly experience text detection using a provided image URL.

```bash
paddleocr text_detection -i https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/general_ocr_001.png
```

--------------------------------

### Output Example

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/pipeline_usage/PP-ChatOCRv4.md

This is the expected output when running the provided Python script.

```text
{'chat_res': {'驾驶室准乘人数': '2'}}
```

--------------------------------

### Install TensorRT wheel package

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/deployment/high_performance_inference.md

Example command to install the TensorRT wheel package from a downloaded tar file.

```bash
python -m pip install TensorRT-8.6.1.6/python/tensorrt-8.6.1-cp310-none-linux_x86_64.whl
```

--------------------------------

### Install PaddlePaddle (GPU with CUDA 12.6)

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/paddlex/quick_start.en.md

Command to install the GPU version of PaddlePaddle, specifying CUDA 12.6.

```bash
python -m pip install paddlepaddle-gpu==3.0.0 -i https://www.paddlepaddle.org.cn/packages/stable/cu126/
```

--------------------------------

### Initialize DocPreprocessor

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/PaddleOCR/docs/version3.x/pipeline_usage/doc_preprocessor.en.md

Example of initializing the DocPreprocessor, specifying GPU usage.

```python
docpp = DocPreprocessor(device="gpu") # Specify whether to use GPU for model inference via device
```

--------------------------------

### Manual Start (Backend and Frontend Separated)

Source: https://github.com/ymylive/chaoxing-fanya/blob/main/README.md

Instructions for starting the backend and frontend separately.

```bash
# 后端
pip install -r requirements.txt
python app.py        # 默认 http://localhost:5000

# 前端（新终端）
cd web
npm install
npm run dev          # 默认 http://localhost:3000
```
Tokens:
Raw