Code for the paper: Solving Linear Inverse Problems using GAN priors
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.
exp_scripts update Feb 26, 2018
src Update Feb 26, 2018

Solving Linear Inverse Problems using GANs

Code for the paper: Solving Linear Inverse Problems Using GAN Priors: An Algorithm with Provable Guarantees.


To run this code, you require Python 2.7, Tensorflow 1.0.1 (preferably with GPU support), Scipy and PyPNG.


  1. Clone the repository, and run all the commands from the parent directory, pgdgan/.

  2. Download the datasets with the script*:

    $ ./setup/ 
  3. To train the DCGAN on celebA from scratch, please visit, and follow the instructions. Else, pretrained GAN model is available, courtesy Bora et al. To download it, please run the following script*:

    $ ./setup/

    Make sure the model is located at ./models/celebA_64_64/

  4. Run following to run the experiment:

    $ python ./src/

    You can also use the script available in ./exp_scripts/

* replicated from .