Text-To-Video-Finetuning
Text-To-Video-Finetuning copied to clipboard
Enabling Multi-GPU training
How to enable multi-GPU training? No matter how many GPUs I use, only one process starts.
I'm also curious about this
Is there any progress?
FYI, since the training code is wrapped in Accelerator
, I just launch training with multiple GPUs and it seems work. Also, make sure that some functions are called only once in the main process by checking accelerator.is_main_process
(e.g., create_output_folders
, save_pipe
, and the evaluation part)