Skip to content
Tensorflow implementation of pixel-recursive-super-resolution(Google Brain paper: https://arxiv.org/abs/1702.00783)
Branch: master
Clone or download
Latest commit 925d04a Jul 2, 2017
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
assets first submit Feb 21, 2017
tools fix path bug Jul 2, 2017
.gitignore first submit Feb 21, 2017
LICENSE add LICENSE Feb 21, 2017
README.md fix readme bug Feb 21, 2017
data.py fix num_epoch bug Mar 15, 2017
net.py first submit Feb 21, 2017
ops.py first submit Feb 21, 2017
solver.py fix num_epoch bug Mar 15, 2017
utils.py first submit Feb 21, 2017

README.md

Pixel Recursive Super Resolution

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

model

Requirements

Usage

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

Samples

Training after 30000 iteration.

sample.png

Training details

cross entropy loss:

curve.png

Author

nilboy / @nilboy

You can’t perform that action at this time.