Quantize_Pytorch_Vgg16AndMobileNet

Quantize,Pytorch,Vgg16,MobileNet

Static Quantize And Aware training Quantize for VGG-16 and MobileNet-V2

Pytorch版本要求：1.4+

Note：VGG-16中没有BN层，所以相较官方教程，去掉了fuse_model的融合部分

项目目录结构：

----Quantize_Pytorch：总项目文件夹

--------data：文件夹，存储imagenet_1k数据集

--------model：文件夹，存储VGG-16以及MobieNet的pretrained_model预训练float型模型

--------MobileNetV2-quantize_all.py：Static Quantize 和 QAT 两种方式

--------vgg16-static_quantize.py：Static Quantize to Vgg-16

--------vgg16-aware_train_quantize.py：QAT Quantize to Vgg-16

数据集与预训练模型下载：链接：https://pan.baidu.com/s/1-Rkrcg0R5DbNdjQ4umXBtQ ；提取码：p2um

Name		Name	Last commit message	Last commit date
Latest commit History 13 Commits
data		data
model		model
quan_bin		quan_bin
quan_bin_pertensor		quan_bin_pertensor
.gitignore		.gitignore
=1.4.0		=1.4.0
MobileNetV2-quantize_all.py		MobileNetV2-quantize_all.py
README.md		README.md
nie.ipynb		nie.ipynb
requirements.txt		requirements.txt
static_quantize.ipynb		static_quantize.ipynb
static_quantize_per_tensor.ipynb		static_quantize_per_tensor.ipynb
vgg16-aware_train_quantize.py		vgg16-aware_train_quantize.py
vgg16-static_quantize.py		vgg16-static_quantize.py