https://github.com/linchuming/imagesr-pytorch

An image super-resolution implemented by Pytorch
https://github.com/linchuming/imagesr-pytorch

Last synced: 2 months ago
JSON representation

An image super-resolution implemented by Pytorch

• Host: GitHub
• URL: https://github.com/linchuming/imagesr-pytorch
• Owner: linchuming
• License: mit
• Created: 2019-11-25T10:57:14.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2019-12-01T12:01:46.000Z (over 5 years ago)
• Last Synced: 2025-04-02T11:11:15.775Z (2 months ago)
• Language: Python
• Size: 1.09 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# ImageSR-Pytorch

[![MIT License](http://img.shields.io/badge/license-MIT-blue.svg?style=flat-square)][license]

[license]: https://github.com/linchuming/ImageSR-Pytorch/blob/master/LICENSE

### Preparation

- Python 3.6

- Pytorch 1.1.0

### Model

- Use a simple version EDSR for image super-resolution

### Training

- Prepare training dataset. Please download DIV2K dataset:

[link](http://data.vision.ee.ethz.ch/cvl/DIV2K/DIV2K_train_HR.zip)

- Move the training dataset to ./DIV2K_train_HR

- Check the config in config.yaml

- Run the script `python train.py config.yaml`

### Inference

- Check the config in config.yaml

- Run the script `python inference.py config.yaml`

### Reference

- [ImageSR for Tensorflow](https://github.com/linchuming/ImageSR-Tensorflow)