https://github.com/bgavran/improved_wgan
Implementation of the "Improved Training of Wasserstein GANs" paper in TensorFlow
Science Score: 36.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
○CITATION.cff file
-
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
✓Academic publication links
Links to: arxiv.org -
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (5.5%) to scientific vocabulary
Repository
Implementation of the "Improved Training of Wasserstein GANs" paper in TensorFlow
Basic Info
Statistics
- Stars: 18
- Watchers: 5
- Forks: 6
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
Improved training of Wasserstein GANs
In this project, the paper Improved training of Wasserstein GANs was implemented in Tensorflow 1.2.0 and Python 3.6.
The paper is the improvement of the Wasserstein GAN paper, which again is the improvement over the original Generative Adversarial Networks paper.
Each of those extension papers represents a step to a more stable training regime.
Improved WGAN, compared to GAN:
- Uses a different distance measure to compare distributions (Wasserstein instead of KL-divergence)
- Enforces the Lipschitz constraint on the critic using gradient penalty
The images on the top are some of the best results with DCGAN and custom upsampling architecture. The dataset used was Labeled Faces in the Wild, the deep-funneled version.
Sample critic and generator training provided below:
This project was created as a part of the FER course Analysis of massive data sets.
Owner
- Name: Bruno Gavranović
- Login: bgavran
- Kind: user
- Location: London, United Kingdom
- Company: Symbolica
- Website: www.brunogavranovic.com
- Repositories: 26
- Profile: https://github.com/bgavran
Principal Scientist - Categorical Deep Learning @symbolica-ai
GitHub Events
Total
Last Year
Issues and Pull Requests
Last synced: about 1 year ago
All Time
- Total issues: 1
- Total pull requests: 0
- Average time to close issues: about 1 year
- Average time to close pull requests: N/A
- Total issue authors: 1
- Total pull request authors: 0
- Average comments per issue: 6.0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Past Year
- Issues: 0
- Pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Issue authors: 0
- Pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Top Authors
Issue Authors
- chunfengshiliburuni (1)