---
title: CycleGAN Flax
emoji: 📚
colorFrom: green
colorTo: yellow
sdk: docker
pinned: false
---
# CycleGAN in Jax + Flax
This space showcases CycleGAN models that were trained on the horse2zebra and monet2photo TensorFlow datasets in Jax with Flax.
Please see the [Colab notebook](https://colab.research.google.com/github/louwjac/CycleGAN-Flax/blob/main/notebooks/inference.ipynb) for an example of how to instantiate the models and load the trained weights.
It includes a Gradio demo that you can use to make your own demo. The model weights are also available on Hugging Face at : [louwjac/CycleGAN-Flax](https://huggingface.co/louwjac/CycleGAN-Flax).
### TensorBoard logs
The TensorBoard logs for the training runs can be viewed at the links below. The logged images are unfortunately not available due to file size constraints.
1. [horse2zebra](https://tensorboard.dev/experiment/5tJZAfBLQkOJ66BUrBVBkQ/)
2. [monet2photo](https://tensorboard.dev/experiment/GZZKiDoBTl6dsgsqRo7rZw/)