目录

A Jittor implementation of Conditional GAN (CGAN)

Dataset

  • MNIST can download by jittor

  • Fashion-MNIST jittor doesn’t provide. You can download by pytorch.

How to use

  • train

    python cgan_jittor.py

    every epoch, it will save a image in img_jittor_xxx folder.

    the model will be saved as d_xxx.pkl and g_xxx.pkl.

  • generate image

    python gen_image.py

    it will generate images in fid_xxx folder.

evaluate

use FID score to evaluate.

provide pytorch implementation.

python fid_score.py
关于

Jittor

14.6 MB
邀请码
    Gitlink(确实开源)
  • 加入我们
  • 官网邮箱:gitlink@ccf.org.cn
  • QQ群
  • QQ群
  • 公众号
  • 公众号

版权所有:中国计算机学会技术支持:开源发展技术委员会
京ICP备13000930号-9 京公网安备 11010802032778号