djcan
djcan
@KaihuaTang Thanks for your reply. Now I have another question. How to evaluate on my own dataset? It seems that during inference process, we need to provide the target which...
> @JincanDeng Please refer to [this instruction] to prepare the SMPL files: https://github.com/Arthur151/ROMP/blob/master/simple_romp/README.md#installation Thanks! I got it.
> > > @JincanDeng Please refer to [this instruction] to prepare the SMPL files: https://github.com/Arthur151/ROMP/blob/master/simple_romp/README.md#installation > > > > > > Thanks! I got it. > > Hello, did you...
> Hi, thanks for the great work. I noticed in LDM implementation they didn't set clip_denoised=False during sampling but you set this flag here. I noticed the sampled latents are...
> @better629 Thank you for your solution. When I comment `accelerator.is_main_process`, I meet another error in `save_model_hook` fuction. Do you have any ideas? `Traceback (most recent call last): File "train_text_to_image_dit.py",...
> i was working on this support but after seeing the results of the model i'm not sure it's ready to be added yet: > >  > > there's...
@bghira I use `""` prompt with `0.1` probability for dropout.
@KaihuaTang Can you release the visualization code for sgdet results?
Same question. Somebody knows the reason?
> Thanks for your advice. Lumina is still under actively development. We will release a diffuser version once we reach a stable version. Looking forward to it!