apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-07 09:31:35 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Quantization] Improve the usage of fastdeploy tools (#722)

Browse Source 
Improve the usage of fastdeploy tools
This commit is contained in:
yunyaoXYY
2022-11-28 16:55:32 +08:00
committed by GitHub
parent 9ae5c24607
commit ae3487560d
10 changed files with 24 additions and 30 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
tools/common_tools/auto_compression/README.md
View File

@@ -40,10 +40,10 @@ wget https://bj.bcebos.com/paddlehub/fastdeploy/COCO_val_320.tar.gz
tar -xvf COCO_val_320.tar.gz
```
##### 2.使用fastdeploy --auto_compress命令，执行一键模型自动化压缩:
##### 2.使用fastdeploy compress命令，执行一键模型自动化压缩:
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 --auto_compress --config_path=./configs/detection/yolov5s_quant.yaml --method='PTQ' --save_dir='./yolov5s_ptq_model/'
fastdeploy compress --config_path=./configs/detection/yolov5s_quant.yaml --method='PTQ' --save_dir='./yolov5s_ptq_model/'
```
##### 3.参数说明
@@ -74,12 +74,12 @@ wget https://bj.bcebos.com/paddlehub/fastdeploy/COCO_train_320.tar
tar -xvf COCO_train_320.tar
```
##### 2.使用fastdeploy --auto_compress命令，执行一键模型自动化压缩:
##### 2.使用fastdeploy compress命令，执行一键模型自动化压缩:
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 --auto_compress --config_path=./configs/detection/yolov5s_quant.yaml --method='QAT' --save_dir='./yolov5s_qat_model/'
fastdeploy compress --config_path=./configs/detection/yolov5s_quant.yaml --method='QAT' --save_dir='./yolov5s_qat_model/'
```
##### 3.参数说明

8
tools/common_tools/auto_compression/README_EN.md
View File

@@ -46,12 +46,12 @@ wget https://bj.bcebos.com/paddlehub/fastdeploy/COCO_val_320.tar.gz
tar -xvf COCO_val_320.tar.gz
```
##### 2. Run fastdeploy --auto_compress command to compress the model
##### 2. Run fastdeploy compress command to compress the model
The following command is to quantize the yolov5s model, if developers want to quantize other models, replace the config_path with other model configuration files in the configs folder.
```shell
fastdeploy --auto_compress --config_path=./configs/detection/yolov5s_quant.yaml --method='PTQ' --save_dir='./yolov5s_ptq_model/'
fastdeploy compress --config_path=./configs/detection/yolov5s_quant.yaml --method='PTQ' --save_dir='./yolov5s_ptq_model/'
```
[notice] PTQ is short for post-training quantization
@@ -83,14 +83,14 @@ wget https://bj.bcebos.com/paddlehub/fastdeploy/COCO_val_320.tar.gz
tar -xvf COCO_val_320.tar.gz
```
##### 2.Use fastdeploy --auto_compress command to compress models
##### 2.Use fastdeploy compress command to compress models
The following command is to quantize the yolov5s model, if developers want to quantize other models, replace the config_path with other model configuration files in the configs folder.
```shell
# Please specify the single card GPU before training, otherwise it may get stuck during the training process.
export CUDA_VISIBLE_DEVICES=0
fastdeploy --auto_compress --config_path=./configs/detection/yolov5s_quant.yaml --method='QAT' --save_dir='./yolov5s_qat_model/'
fastdeploy compress --config_path=./configs/detection/yolov5s_quant.yaml --method='QAT' --save_dir='./yolov5s_qat_model/'
```
##### 3.Parameters

0
tools/common_tools/auto_compression/__init__.py Normal file
View File

1
tools/common_tools/auto_compression/fd_auto_compress/__init__.py
View File

@@ -1 +0,0 @@
import fd_auto_compress

2
tools/common_tools/auto_compression/fd_auto_compress/fd_auto_compress.py
View File

@@ -22,7 +22,7 @@ import paddle
from paddleslim.common import load_config, load_onnx_model
from paddleslim.auto_compression import AutoCompression
from paddleslim.quant import quant_post_static
from fd_auto_compress.dataset import *
from .dataset import *
def argsparser():