https://github.com/hsm207/my_unit
My implementation of the domain adaptation section of the "Unsupervised Image-to-Image Translation Networks" paper
https://github.com/hsm207/my_unit
Last synced: 7 months ago
JSON representation
My implementation of the domain adaptation section of the "Unsupervised Image-to-Image Translation Networks" paper
- Host: GitHub
- URL: https://github.com/hsm207/my_unit
- Owner: hsm207
- Created: 2018-01-09T23:59:22.000Z (almost 8 years ago)
- Default Branch: models
- Last Pushed: 2018-02-08T01:17:33.000Z (over 7 years ago)
- Last Synced: 2025-01-14T13:52:30.755Z (9 months ago)
- Language: Python
- Homepage:
- Size: 43.9 KB
- Stars: 0
- Watchers: 4
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# my_unit
My reimplementation of the SHVN -> MNIST domain adaptation model as described in the 2017 NIPS paper titled [Unsupervised Image-to-Image Translation Networks](https://arxiv.org/pdf/1703.00848.pdf).
This implementation is in TensorFlow and is based on the author's [PyTorch implementation](https://github.com/mingyuliutw/UNIT).
# Status
### 27 January 2018
I have trained the model for 909 epochs and the highest accuracy is around 13%.