https://github.com/the-database/trainner-redux
Deep learning training framework for image super resolution and restoration.
https://github.com/the-database/trainner-redux
deep-learning image-restoration machine-learning neural-network python pytorch super-resolution upscale
Last synced: about 1 year ago
JSON representation
Deep learning training framework for image super resolution and restoration.
- Host: GitHub
- URL: https://github.com/the-database/trainner-redux
- Owner: the-database
- License: apache-2.0
- Created: 2024-06-02T06:24:38.000Z (about 2 years ago)
- Default Branch: master
- Last Pushed: 2025-03-12T20:57:08.000Z (about 1 year ago)
- Last Synced: 2025-03-12T22:00:50.653Z (about 1 year ago)
- Topics: deep-learning, image-restoration, machine-learning, neural-network, python, pytorch, super-resolution, upscale
- Language: Python
- Homepage: https://trainner-redux.readthedocs.io
- Size: 66 MB
- Stars: 50
- Watchers: 1
- Forks: 8
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE.txt
Awesome Lists containing this project
README
# traiNNer-redux
## Overview
[traiNNer-redux](https://trainner-redux.readthedocs.io/en/latest/index.html) is a deep learning training framework for image super resolution and restoration which allows you to train PyTorch models for upscaling and restoring images and videos. NVIDIA graphics card is recommended, but AMD works on Linux machines with ROCm.
## Usage Instructions
Please see the [getting started](https://trainner-redux.readthedocs.io/en/latest/getting_started.html) page for instructions on how to use traiNNer-redux.
## Contributing
Please see the [contributing](https://trainner-redux.readthedocs.io/en/latest/contributing.html) page for more info on how to contribute.
## Resources
- [OpenModelDB](https://openmodeldb.info/): Repository of AI upscaling models, which can be used as pretrain models to train new models. Models trained with this repo can be submitted to OMDB.
- [chaiNNer](https://github.com/chaiNNer-org/chaiNNer): General purpose tool for AI upscaling and image processing, models trained with this repo can be run on chaiNNer. chaiNNer can also assist with dataset preparation.
- [WTP Dataset Destroyer](https://github.com/umzi2/wtp_dataset_destroyer): Tool to degrade high quality images, which can be used to prepare the low quality images for the training dataset.
- [Video Destroyer](https://github.com/Kim2091/video-destroyer): A tool designed to create datasets from a single high quality input video for AI training purposes.
- [helpful-scripts](https://github.com/Kim2091/helpful-scripts): Collection of scripts written to improve experience training AI models.
- [Enhance Everything! Discord Server](https://discord.gg/cpAUpDK): Get help training a model, share upscaling results, submit your trained models, and more.
- [vs_align](https://github.com/pifroggi/vs_align): Video Alignment and Synchonization for Vapoursynth, tool to align LR and HR datasets.
- [ImgAlign](https://github.com/sonic41592/ImgAlign): Tool for auto aligning, cropping, and scaling HR and LR images for training image based neural networks.
- [Deep Learning Tuning Playbook](https://developers.google.com/machine-learning/guides/deep-learning-tuning-playbook): Document to help you train deep learning models more effectively.
## License and Acknowledgement
traiNNer-redux is released under the [Apache License 2.0](LICENSE.txt). See [LICENSE](LICENSE/README.md) for individual licenses and acknowledgements.
- This repository is a fork of [joeyballentine/traiNNer-redux](https://github.com/joeyballentine/traiNNer-redux) which itself is a fork of [BasicSR](https://github.com/XPixelGroup/BasicSR).
- Network architectures are imported from [Spandrel](https://github.com/chaiNNer-org/spandrel).
- Several architectures are developed by [umzi2](https://github.com/umzi2): [ArtCNN-PyTorch](https://github.com/umzi2/ArtCNN-PyTorch), [DUnet](https://github.com/umzi2/DUnet), [FlexNet](https://github.com/umzi2/FlexNet), [MetaGan](https://github.com/umzi2/MetaGan), [MoESR](https://github.com/umzi2/MoESR), [MoSR](https://github.com/umzi2/MoSR), [RTMoSR](https://github.com/rewaifu/RTMoSR), [SPANPlus](https://github.com/umzi2/SPANPlus)
- The [ArtCNN](https://github.com/Artoriuz/ArtCNN) architecture is originally developed by [Artoriuz](https://github.com/Artoriuz).
- The TSCUNet architecture is from [aaf6aa/SCUNet](https://github.com/aaf6aa/SCUNet) which is a modification of [SCUNet](https://github.com/cszn/SCUNet), and parts of the training code for TSCUNet are adapted from [TSCUNet_Trainer](https://github.com/Demetter/TSCUNet_Trainer).
- Several enhancements reference implementations from [Corpsecreate/neosr](https://github.com/Corpsecreate/neosr) and its original repo [neosr](https://github.com/muslll/neosr).
- Members of the Enhance Everything Discord server: [Corpsecreate](https://github.com/Corpsecreate), [joeyballentine](https://github.com/joeyballentine), [Kim2091](https://github.com/Kim2091).