huan ling

Maybe you want to make the editing region smaller to test first. Once you can deploy the model, the memory depends on your editing region area.

what's your cuda version and system env?

I don't think cuda 11.5.1 is supported. The code is tested with CUDA 10.1 toolkit with Pytorch==1.4.0 and CUDA 11.4 with Pytorch==1.10.0. Check pytorch install doc at https://pytorch.org/get-started/previous-versions/

Thanks! > I think the torch version in the requirements is an old alpha version. The version number has that **a** in it for alpha. Maybe old alphas aren't available?...

why do you need to use ninja -v? It's not in required package

You need to host front end on the cloud server as well. Could you open chrome's inspect and post the error message here?

That's the latent vector generated automatically. Maybe create the folder "./static/results/" manually ?

Hey we didn't provide deeplab checkpoints.

Hi, Thank you for pointing this out. There is a discrepancy caused by the noise channel and we are looking into it. At the same time, please ignore those two...

Hi, we pause sharing code because we plan to release a new platform with lots of different components to the community. Coming soon. Stay tuned. Thank you for your patience....