zhenglin 1.20.20

A deep-learning package contains a set of well-organized deep-neural networks&tools.

pip install zhenglin

@ Author: zhenglin @version: 1.20.19

This package contains some off-the-shelves deep-learning networks implemented with .

use

pip install zhenglin

to install this package.

News and Updates

• Aug 13 2023: Version: 1.18.15 Add basic quantization support.
• Oct 26 2023: Version: 1.20.19 Add TransUNet and SwinUNet

Introduction

zhenglin package is mainly motivated by eriklindernoren and his repo which provides many super clean and easy-to-read implementation of GAN variants. It is friendly to beginners and also a good reference for advanced users, especially for DL developpers.

Specifically, this package provides

• A universal structure under zhenglin.dl.template.v1.*
• Loss functions under zhenglin.dl.losses
• Metrics under zhenglin.dl.metrics
• Utils under zhenglin.dl.utils
• 20 highly modular and easy-to-use implementation of deep-learning networks under zhenglin.dl.networks.* which includes(from a to z)

• cycleGAN
• DDPM
• DeblurGAN
• EDSR
• ESRGAN
• Noise2Void
• Pix2Pix
• RCAN
• ResNet
• Restormer
• RRDBNet
• SqueezeNet
• SRDRM
• SRGAN
• SwinIR
• SwinUNet
• TransUNet
• U2Net
• UNet
• Attention-UNet