# BGAN-TF

**Repository Path**: zyabo/bgan-tf

## Basic Information

- **Project Name**: BGAN-TF
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: MIT
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2020-12-22
- **Last Updated**: 2020-12-22

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# bgan-tf
Implementing the Boundary-Seeking GAN in TensorFlow. Credit to the authors of the [original paper](https://arxiv.org/abs/1702.08431v2), who also released their [Theano implementation](https://github.com/rdevon/BGAN).

#### Usage

First, download the celebrity faces (CelebA) dataset, put it in `data/`, and unzip it such that we have `data/img_align_celeba/`.

Then prepare the tfrecords:

```
python make_tfrecords.py
```

Set training params in train.py and run training with:

```
python train.py
```