Conditional CoopNets

This repository contains a TensorFlow implementation for TPAMI paper "Cooperative Training of Fast Thinking Initializer and Slow Thinking Solver for Conditional Learning"

Set Up Environment

We have provided the environment file cCoopNets.yml for setting up the environment. The environment can be set up with one command using conda

conda env create -f cCoopNets.yml
conda activate cCoopNets

Reference

@article{DG,
    author = {Jianwen Xie, Zilong Zheng, Xiaolin Fang, Song-Chun Zhu, Ying Nian Wu},
    title = {Cooperative Training of Fast Thinking Initializer and Slow Thinking Solver for Conditional Learning},
    journal={IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
    year = {2021}
}

Training

Conditional image generation on grayscale images

(a) MNIST dataset

$ python main_mnist.py --category MNIST

(b) fashion-MNIST dataset

$ python main_mnist.py --category fashion_MNIST