Did you already ask in the discord?: Yes
You verified that this is a bug and not a feature request or question by asking in the discord?: Yes, pretty sure
When I run "python run.py train_lora_flux_24gb.yaml" where "train_lora_flux_24gb.yaml" is edited the way it should be (I'm running the training on my other computer, but that's not running on the absolutely latest git pull) it gets stuck downloading models. First I got a message about symlinks not being supported, but deactivating it using "set HF_HUB_DISABLE_SYMLINKS=1" before running the training it still gets stuck downloading.
Is there any way to manually download all files and run the training pointing to a specific folder or model directly instead of just relying on the huggingface cache-download? That would have been so much easier!
This is for bugs only
Did you already ask in the discord?: Yes You verified that this is a bug and not a feature request or question by asking in the discord?: Yes, pretty sure
When I run "python run.py train_lora_flux_24gb.yaml" where "train_lora_flux_24gb.yaml" is edited the way it should be (I'm running the training on my other computer, but that's not running on the absolutely latest git pull) it gets stuck downloading models. First I got a message about symlinks not being supported, but deactivating it using "set HF_HUB_DISABLE_SYMLINKS=1" before running the training it still gets stuck downloading.
Is there any way to manually download all files and run the training pointing to a specific folder or model directly instead of just relying on the huggingface cache-download? That would have been so much easier!