TensorFlow implementation of Pixel Recursive Super Resolution. This implementation contains:

• Python 2.7
• Skimage
• TensorFlow 1.0

First, download data celebA

$ mkdir data
$ cd data
$ ln -s $celebA_path celebA

Then, create image_list file:

$ python tools/create_img_lists.py --dataset=data/celebA --outfile=data/train.txt

To train model on gpu:

$ python tools/train.py
(or $ python tools/train.py --device_id=0)

To train model on cpu: $ python tools/train.py --use_gpu=False

Training after 30000 iteration.

cross entropy loss:

nilboy / @nilboy