# cycleGAN-keras

**Repository Path**: yang20072845/cycleGAN-keras

## Basic Information

- **Project Name**: cycleGAN-keras
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2020-07-02
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# cycleGAN-keras

Fully working keras implementation of cycleGAN in Python, realized during internship at University of Parma's <a  href="http://implab.ce.unipr.it"> IMPLab </a>.

  

\[<a  href="https://github.com/junyanz/CycleGAN"> CycleGAN Authors' repository</a> - <a  href="https://arxiv.org/pdf/1703.10593.pdf">CycleGAN paper</a>  \]


<h2>Usage</h2>

To download toy example datasets (check available datasets <a href="">here</a>) use the datasets script:

`sh datasets.sh dataset_name`  <br>

To enable faster image loading into the network you can use hdf5 format by setting `use_hdf5=True` in `train` function in `cycleGAN.py`. <br>

To build `.hdf5` datsets use the `make_dataset` script in datasets folder:

`python3 make_dataset.py dataset_path number_of_channels`  <br>

For those with limited GPU resources, I suggest downscaling the images through the `--img_size` argument (square images are input and output of the network). <br>

To <b>train</b> the network: `python3 cycleGAN.py` <br>

At the end of each epoch a sample image of predictions on test images is output into `./images` folder, along with the losses plots.
  
<h2>Dependencies</h2>

<ul>

<li>Python 3</li>

<li>Tensorflow 1.2 or greater</li>

<li>Keras 2.1.2</li>

<li>keras-contrib</li>

<li>h5py</li>

<li>glob</li>

<li>tqdm</li>

</ul>


<h2>Samples</h2>

<h3>horse2zebra</h3>

![144-horse2zebra](https://user-images.githubusercontent.com/24715359/48908870-ea066880-ee6b-11e8-905c-47c4278f1c14.png)

<h3>cityscapes</h3>

![198_cityscapes_resnet](https://user-images.githubusercontent.com/24715359/48908826-c511f580-ee6b-11e8-9669-b58685e7171f.png)

<h3>maps</h3>

![195_maps_resnet](https://user-images.githubusercontent.com/24715359/48908829-c6432280-ee6b-11e8-9cf0-cc3286878fa6.png)

<h2>Credits</h2>
Keras residual block implementation by <a href="https://github.com/relh">Richard Higgins</a>