Arka Sadhu

I can confirm as well. It now uses 1.7.6. I am not sure what changed though. My conda env is same as it was last time. My guess is some...

I am having the same issue. Is there something which breaks in python 3.9 in particular? If not, would it be possible to to support python 3.9? Thanks!

@rom1504 Yes, img2dataset is awesome and I used it to download cc3m. I am preparing a webdataset on my end using tarp for a video dataset so I was wondering...

@rom1504 You are right. With 3gb shards, for simple dataloading I see 2x speed up if I just read the video (don't decode it), but just 5% speedup if I...

Did you try this out @lxtGH? If so can you share the performance?

Has anyone been able to figure it out?

Hi @Leo-Yuyang Sorry, this completely escaped me. I am more easily reached over email ([email protected]). Were you able to download it? I believe the google drive download code may not...

@TheLastBen If I understand correctly, this file converts to half precision as well (https://github.com/TheLastBen/fast-stable-diffusion/blob/main/Dreambooth/convertosd.py#L226) when saving the final checkpoint. For my use case, I have a local GPU that only...

@kratzert I am not sure why, your method didn't work for me. What worked for me instead is this ``` import sys sys.path.append(path_to_root_of_this_repo) import kaffe.tensorflow ``` Equivalently, one can add...

Hi @pinkfloyd06. I am not the author but I have read the code and I hope that I could answer some of those questions. Do note I too might have...