CycleGAN-TensorFlow

An implementation of CycleGan using TensorFlow (work in progess).

Original paper: https://arxiv.org/abs/1703.10593

Environment

• TensorFlow 1.0.0
• Python 3.6.0

Data preparing

• First, download a dataset, e.g. apple2orange

$ bash download_dataset.sh apple2orange

• Write the dataset to tfrecords

$ python dump.py

Training

$ python train.py

Check TensorBoard to see training progress and generated images.

$ tensorboard --logdir checkpoints/${datetime}

• apple -> orange

• orange -> apple

TODO:

• Instance normalization
• Update discriminators using a history of generated images