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...

You can try our HairCLIP on [Replicate](https://replicate.com/wty-ustc/hairclip), it does not require you to have ML background to experience it.

I'm not sure which step you're having trouble with, can you be more specific?

Thanks for the suggestion, I have updated the readme, please check it out. Also, predict.py and cog.yaml are for Replicate only.

No. You can get the dataset from this [repository](https://github.com/tkarras/progressive_growing_of_gans).

The hairstyle_ref_img_train_path needs to be provided with images. Can you provide more information about the error?

No ---Original--- From: "Muhammed ***@***.***> Date: Sat, Feb 19, 2022 23:10 PM To: ***@***.***>; Cc: ***@***.******@***.***>; Subject: Re: [dukebw/LOHO] Code for inference on my own data (Issue #7) Hi @wty-ustc,...

In our current version, text proxy, reference proxy, and color proxy are generated by optimization, and sketch proxy is generated by the forward pass. Do you mean how the sketch...

I guess he wasn't correctly aligned.

Sorry, there is no pre-trained model on this dataset, you can train one yourself. However, my personal experience is that the model pre-trained on FFHQ performs well enough.