AfterHAL

Hi. Same error for me. Running on RX4080 16GB Vram (2GB used by system)  GPU VRAM is filled up in 10 seconds. 

Hi. It only works with absolute path for me. In my case, on W: drive : W:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-IDM-VTON\models\idm_vton

Hi there. Does the regularization files for Flux training needs captioning ?

> Good one. Thanks man. "&ref=2pdhmpu1" in the URL --> Affiliation Template Referral Program with 1% Commission. mp3pintyo should be the one who says "thank you".

> > > Good one. Thanks man. > > > > > > "&ref=2pdhmpu1" in the URL --> Affiliation Template Referral Program with 1% Commission. > > mp3pintyo should be...

> OH, this is because we save the seed and randomize information in the checkpoint. So when you resume from a ckpt saved on a 7-GPUs node and run the...

> Oh, it should be --model.load_from xxx.pth Hi. I already tried with `--model.load_from xxx.pth`, but it gives another error: ``` [rank0]: File "/mnt/d/TODAI/apps/SanaLinux/Sana/train_scripts/train.py", line 798, in main [rank0]: load_ema=config.model.resume_from.get("load_ema", False),...

Hi. I'm still stuck with this problem and have no clue... Can someone help me please?

Hi @lawrence-cj . I'm still stuck with it. Some helps will be appreciated. Thanks.

Thank you @lawrence-cj . This checkpoint is currently at 222k steps trained on 2 GPUs. Here is the command to (re)start this training session on 1 GPU, after manually changing...