search

znxlwm / pytorch-MNIST-CelebA-GAN-DCGAN

Pytorch implementation of Generative Adversarial Networks (GAN) and Deep Convolutional Generative Adversarial Networks (DCGAN) for MNIST and CelebA datasets
511 stars 145 forks source link

why is the size of generated mnist image is (64,64) rather than (28,28)? #3

Open zhangrong1722 opened 6 years ago

zhangrong1722 commented 6 years ago

I check your DCGAN implementation details.Why is the size of generated mnist image is (64,64) rather than (28,28)?

thanks.

znxlwm commented 5 years ago

Hi,

Because DCGAN paper used 64 x 64. so I just used 64 x 64.

You can find more details in DCGAN paper ( https://arxiv.org/pdf/1511.06434.pdf).

2018년 11월 20일 (화) 오후 12:58, taozhuang123 notifications@github.com님이 작성:

if you know why,please teach me,very thanks

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/znxlwm/pytorch-MNIST-CelebA-GAN-DCGAN/issues/3#issuecomment-440129645, or mute the thread https://github.com/notifications/unsubscribe-auth/AGt-hhBOeTTuMycZ6MsX5aV_Yw34WUSOks5uw33LgaJpZM4UW0kc .

--

강현우 Hyeonwoo Kang

Vision TF, AI Center

12, Daewangpangyo-ro 644beon-gil, Bundang-gu, Seongnam-si, Gyeonggi-do, Korea

Tel 82 2 6201 0076

Mobile 82 10 5291 6774

Email hwkang0131@ncsoft.com pathos@ncsoft.com