Issues about switch off the update G

[wtybest]

When I switch off the options --update_G, to perform the reconstruction, which means only optimize the latent code, the result seems far from satisfactory, even for imagenet val datasets.

In my understanding, optimizing just for latent code might not be perfect, but at least it shouldn't be this bad.
Is that consistent with what you tried?

[XingangPan]

@wty-ustc Yes, the result is consistent with what I tried.
Optimizing just for latent code would produce very inaccurate reconstruction for BigGAN, even if you enlarge the learning rate. That's why we need to optimize the generator too.

[wtybest]

Thank you for your response! 

…

------------------ 原始邮件 ------------------ 发件人: "XingangPan/deep-generative-prior" ***@***.***>; 发送时间: 2021年4月1日(星期四) 下午3:05 ***@***.***>; ***@***.******@***.***>; 主题: Re: [XingangPan/deep-generative-prior] Issues about switch off the update G (#10) @wty-ustc Yes, the result is consistent with what I tried. Optimizing just for latent code would produce very inaccurate reconstruction for BigGAN, even if you enlarge the learning rate. That's why we need to optimize the generator too. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[zhangxiangyu19]

it's the same with my test. How do other generators do it, is it because the training set is smaller?