search

imlixinyang / HiSD

Code for "Image-to-image Translation via Hierarchical Style Disentanglement" (CVPR 2021 Oral).
Other
392 stars 47 forks source link

A little problem about the batchsize #20

Closed xiaolongcheng closed 3 years ago

xiaolongcheng commented 3 years ago

Hi, your work is so impressive that I want to reproduce your experiment. But I got a problem about the config of the batchsize, I use a single GTX 1080Ti GPU as you did，but the batchsize of 8 seems too big for it, I can just set batchsize to 5 or 4. Can you tell me the solution about this problem,thanks a lot

imlixinyang commented 3 years ago

Which config file did you use?

xiaolongcheng commented 3 years ago

Oh! This do remind me about the configs, I use celena-hq_256.yaml as my configuration. If I want to set batchsize to 8, I should use celebs-hq.yaml, which set the input_size as 128, am I right?

------------------ 原始邮件 ------------------ 发件人: "imlixinyang/HiSD" @.>; 发送时间: 2021年6月3日(星期四) 晚上6:06 @.>; @.**@.>; 主题: Re: [imlixinyang/HiSD] A little problem about the batchsize (#20)

Which config file did you use?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

imlixinyang commented 3 years ago

Yes. The 256 config needs 2 1080ti at least.

xiaolongcheng commented 3 years ago

Thanks a lot for your time and reply

---Original--- From: @.> Date: Thu, Jun 3, 2021 19:45 PM To: @.>; Cc: @.**@.>; Subject: Re: [imlixinyang/HiSD] A little problem about the batchsize (#20)

Yes. The 256 config needs 2 1080ti at least.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

imlixinyang commented 3 years ago

Always welcomed!