PAGER: Progressive Attribute-Guided Extendable Robust image generation

This repo containes codes for the following paper:

@article{azizi2022pager,
  title={PAGER: Progressive Attribute-Guided Extendable Robust Image Generation},
  author={Azizi, Zohreh and Kuo, C-C Jay and others},
  journal={APSIPA Transactions on Signal and Information Processing},
  volume={11},
  number={1},
  year={2022},
  publisher={Now Publishers, Inc.}
}

Installation

• Run the folowing lines to create a conda environment:

conda init bash
source ~/.bashrc
conda create -n PAGER python==3.8.5
conda activate PAGER

• Run bash install.sh to install required packages.

Dataset preparation

The code is tested for MNIST, Fashion-MNIST and CelebA. You must download the datasets and place them in their corresponding folders in the datasets folder.

Usage

• If you use USC HPC server, use the following commands to setup a GPU node:

salloc --time=48:00:00 --ntasks=1 --cpus-per-task=1 --mem=180GB --gres=gpu:v100:1 --partition=gpu

• Load the following cuda and cudnn versions and setup the environment:

conda deactivate
module load gcc/8.3.0
module load cuda/10.1.243
module load cudnn/7.6.5.32-10.1
module load anaconda3
conda activate PAGER

• Run main_celeba.py to train and generate for CelebA dataset.
• Run main_mnist_fashion.py to train and generate for mnist or fashion-mnist datasets.