mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2025-10-30 03:22:05 +08:00
b0663209f6
* Add PaddleOCR Support * Add PaddleOCR Support * Add PaddleOCRv3 Support * Add PaddleOCRv3 Support * Update README.md * Update README.md * Update README.md * Update README.md * Add PaddleOCRv3 Support * Add PaddleOCRv3 Supports * Add PaddleOCRv3 Suport * Fix Rec diff * Remove useless functions * Remove useless comments * Add PaddleOCRv2 Support * Add PaddleOCRv3 & PaddleOCRv2 Support * remove useless parameters * Add utils of sorting det boxes * Fix code naming convention * Fix code naming convention * Fix code naming convention * Fix bug in the Classify process * Imporve OCR Readme * Fix diff in Cls model * Update Model Download Link in Readme * Fix diff in PPOCRv2 * Improve OCR readme * Imporve OCR readme * Improve OCR readme * Improve OCR readme * Imporve OCR readme * Improve OCR readme * Fix conflict * Add readme for OCRResult * Improve OCR readme * Add OCRResult readme * Improve OCR readme * Improve OCR readme * Add Model Quantization Demo * Fix Model Quantization Readme * Fix Model Quantization Readme * Add the function to do PTQ quantization * Improve quant tools readme * Improve quant tool readme * Improve quant tool readme * Add PaddleInference-GPU for OCR Rec model * Add QAT method to fastdeploy-quantization tool * Remove examples/slim for now * Move configs folder * Add Quantization Support for Classification Model * Imporve ways of importing preprocess * Upload YOLO Benchmark on readme * Upload YOLO Benchmark on readme * Upload YOLO Benchmark on readme * Improve Quantization configs and readme * Add support for multi-inputs model * Add backends and params file for YOLOv7 * Add quantized model deployment support for YOLO series * Fix YOLOv5 quantize readme * Fix YOLO quantize readme * Fix YOLO quantize readme * Improve quantize YOLO readme * Improve quantize YOLO readme * Improve quantize YOLO readme * Improve quantize YOLO readme * Improve quantize YOLO readme * Fix bug, change Fronted to ModelFormat * Change Fronted to ModelFormat * Add examples to deploy quantized paddleclas models * Fix readme * Add quantize Readme * Add quantize Readme * Add quantize Readme * Modify readme of quantization tools * Modify readme of quantization tools * Improve quantization tools readme * Improve quantization readme * Improve PaddleClas quantized model deployment readme * Add PPYOLOE-l quantized deployment examples * Improve quantization tools readme
108 lines
4.6 KiB
Markdown
108 lines
4.6 KiB
Markdown
# FastDeploy 一键模型量化
|
||
FastDeploy基于PaddleSlim, 给用户提供了一键模型量化的工具, 支持离线量化和量化蒸馏训练.
|
||
本文档以Yolov5s为例, 供用户参考如何安装并执行FastDeploy的一键模型量化.
|
||
|
||
## 1.安装
|
||
|
||
### 环境依赖
|
||
|
||
1.用户参考PaddlePaddle官网, 安装develop版本
|
||
```
|
||
https://www.paddlepaddle.org.cn/install/quick?docurl=/documentation/docs/zh/develop/install/pip/linux-pip.html
|
||
```
|
||
|
||
2.安装paddleslim-develop版本
|
||
```bash
|
||
git clone https://github.com/PaddlePaddle/PaddleSlim.git & cd PaddleSlim
|
||
python setup.py install
|
||
```
|
||
|
||
### FastDeploy-Quantization 安装方式
|
||
用户在当前目录下，运行如下命令:
|
||
```
|
||
python setup.py install
|
||
```
|
||
|
||
## 2.使用方式
|
||
|
||
### 一键量化示例
|
||
|
||
#### 离线量化
|
||
|
||
##### 1. 准备模型和Calibration数据集
|
||
用户需要自行准备待量化模型与Calibration数据集.
|
||
本例中用户可执行以下命令, 下载待量化的yolov5s.onnx模型和我们为用户准备的Calibration数据集示例.
|
||
|
||
```shell
|
||
# 下载yolov5.onnx
|
||
wget https://paddle-slim-models.bj.bcebos.com/act/yolov5s.onnx
|
||
|
||
# 下载数据集, 此Calibration数据集为COCO val2017中的前320张图片
|
||
wget https://bj.bcebos.com/paddlehub/fastdeploy/COCO_val_320.tar.gz
|
||
tar -xvf COCO_val_320.tar.gz
|
||
```
|
||
|
||
##### 2.使用fastdeploy_quant命令，执行一键模型量化:
|
||
以下命令是对yolov5s模型进行量化, 用户若想量化其他模型, 替换config_path为configs文件夹下的其他模型配置文件即可.
|
||
```shell
|
||
fastdeploy_quant --config_path=./configs/detection/yolov5s_quant.yaml --method='PTQ' --save_dir='./yolov5s_ptq_model/'
|
||
```
|
||
|
||
##### 3.参数说明
|
||
|
||
目前用户只需要提供一个定制的模型config文件,并指定量化方法和量化后的模型保存路径即可完成量化.
|
||
|
||
| 参数 | 作用 |
|
||
| -------------------- | ------------------------------------------------------------ |
|
||
| --config_path | 一键量化所需要的量化配置文件.[详解](./configs/README.md) |
|
||
| --method | 量化方式选择, 离线量化选PTQ，量化蒸馏训练选QAT |
|
||
| --save_dir | 产出的量化后模型路径, 该模型可直接在FastDeploy部署 |
|
||
|
||
|
||
|
||
#### 量化蒸馏训练
|
||
|
||
##### 1.准备待量化模型和训练数据集
|
||
FastDeploy目前的量化蒸馏训练，只支持无标注图片训练，训练过程中不支持评估模型精度.
|
||
数据集为真实预测场景下的图片，图片数量依据数据集大小来定，尽量覆盖所有部署场景. 此例中，我们为用户准备了COCO2017验证集中的前320张图片.
|
||
注: 如果用户想通过量化蒸馏训练的方法,获得精度更高的量化模型, 可以自行准备更多的数据, 以及训练更多的轮数.
|
||
|
||
```shell
|
||
# 下载yolov5.onnx
|
||
wget https://paddle-slim-models.bj.bcebos.com/act/yolov5s.onnx
|
||
|
||
# 下载数据集, 此Calibration数据集为COCO2017验证集中的前320张图片
|
||
wget https://bj.bcebos.com/paddlehub/fastdeploy/COCO_val_320.tar.gz
|
||
tar -xvf COCO_val_320.tar.gz
|
||
```
|
||
|
||
##### 2.使用fastdeploy_quant命令，执行一键模型量化:
|
||
以下命令是对yolov5s模型进行量化, 用户若想量化其他模型, 替换config_path为configs文件夹下的其他模型配置文件即可.
|
||
```shell
|
||
# 执行命令默认为单卡训练，训练前请指定单卡GPU, 否则在训练过程中可能会卡住.
|
||
export CUDA_VISIBLE_DEVICES=0
|
||
fastdeploy_quant --config_path=./configs/detection/yolov5s_quant.yaml --method='QAT' --save_dir='./yolov5s_qat_model/'
|
||
```
|
||
|
||
##### 3.参数说明
|
||
|
||
目前用户只需要提供一个定制的模型config文件,并指定量化方法和量化后的模型保存路径即可完成量化.
|
||
|
||
| 参数 | 作用 |
|
||
| -------------------- | ------------------------------------------------------------ |
|
||
| --config_path | 一键量化所需要的量化配置文件.[详解](./configs/README.md)|
|
||
| --method | 量化方式选择, 离线量化选PTQ，量化蒸馏训练选QAT |
|
||
| --save_dir | 产出的量化后模型路径, 该模型可直接在FastDeploy部署 |
|
||
|
||
|
||
## 3. FastDeploy 部署量化模型
|
||
用户在获得量化模型之后，即可以使用FastDeploy进行部署, 部署文档请参考:
|
||
具体请用户参考示例文档:
|
||
- [YOLOv5 量化模型部署](../../examples/vision/detection/yolov5/quantize/)
|
||
|
||
- [YOLOv6 量化模型部署](../../examples/vision/detection/yolov6/quantize/)
|
||
|
||
- [YOLOv7 量化模型部署](../../examples/vision/detection/yolov7/quantize/)
|
||
|
||
- [PadddleClas 量化模型部署](../../examples/vision/classification/paddleclas/quantize/)
|