multi-gpu training about text-to-video-finetuning HOT 3 CLOSED

Yes，as your suggestion above, i can use the distributed training and the generation is as good as one GPU

from text-to-video-finetuning.

will it be difficult to modify the code to support multi-gpu training?

I've never tried multiple GPU training, but you may be able to do it naively with accelerate.

accelerate config

You should be prompted to configure your setup, including multiple GPU training.
Then it should be as simple as:

accelerate launch train.py --config ./configs/my_config_hq.yaml

Let us know how it goes if you decide to try! If it doesn't I could try to implement it, but I don't have multiple GPUs and would probably need to rent out a server to do so.

from text-to-video-finetuning.

Yes，as your suggestion above, i can use the distributed training and the generation is as good as one GPU

What do you mean with "as good as one gpu"? You mean, if you had a single gpu as powerful as both of the gpus that you are using combined, there would be no difference? Also, are you connecting them through nvlink, or its just pcie?

from text-to-video-finetuning.