Skip to content

This is a practice to make Generative Adversial Network by using google colab.

Notifications You must be signed in to change notification settings

LeeChungHyun/GAN-using-Mnist_data

Repository files navigation

GAN-using-Mnist_data

This is a practice to make Generative Adversial Network(GAN) by using MNist data in google colab.

Grayscale버전은 Mnist_gan_2으로 실험해봤습니다. gan_2에서는 목적 함수를 MSE 대신 cross entopy를 사용했습니다. 따라서 실험중 그레디언트(Loss)를 보면, 오류가 더 큰 쪽에 더 큰 벌점을 부과하는 경향이 보입니다.

실험 내용은 ppt에 첨부했습니다. 감사합니다!

About

This is a practice to make Generative Adversial Network by using google colab.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published