https://github.com/lucidrains/transganformer

Implementation of TransGanFormer, an all-attention GAN that combines the finding from the recent GanFormer and TransGan paper
https://github.com/lucidrains/transganformer

artificial-intelligence attention-mechanism deep-learning generative-adversarial-networks transformers

Last synced: about 11 hours ago
JSON representation

Implementation of TransGanFormer, an all-attention GAN that combines the finding from the recent GanFormer and TransGan paper

• Host: GitHub
• URL: https://github.com/lucidrains/transganformer
• Owner: lucidrains
• License: mit
• Created: 2021-03-11T03:31:10.000Z (over 3 years ago)
• Default Branch: main
• Last Pushed: 2021-04-27T15:14:45.000Z (over 3 years ago)
• Last Synced: 2024-08-08T18:56:34.395Z (about 2 months ago)
• Topics: artificial-intelligence, attention-mechanism, deep-learning, generative-adversarial-networks, transformers
• Language: Python
• Homepage:
• Size: 292 KB
• Stars: 154
• Watchers: 16
• Forks: 15
• Open Issues: 5
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
## TransGanFormer (wip)

Implementation of TransGanFormer, an all-attention GAN that combines the finding from the recent GansFormer and TransGan paper. It will also contain a bunch of tricks I have picked up building transformers and GANs for the last year or so, including efficient linear attention and pixel level attention.

## Install

```bash

$ pip install transganformer

```

## Usage

```bash

$ transganformer --data ./path/to/data

```

## Citations

```bibtex

@misc{jiang2021transgan,

    title   = {TransGAN: Two Transformers Can Make One Strong GAN}, 

    author  = {Yifan Jiang and Shiyu Chang and Zhangyang Wang},

    year    = {2021},

    eprint  = {2102.07074},

    archivePrefix = {arXiv},

    primaryClass = {cs.CV}

}

```

```bibtex

@misc{hudson2021generative,

    title   = {Generative Adversarial Transformers}, 

    author  = {Drew A. Hudson and C. Lawrence Zitnick},

    year    = {2021},

    eprint  = {2103.01209},

    archivePrefix = {arXiv},

    primaryClass = {cs.CV}

}

```