About Training

[XuWuLingYu]

In video training,only unet's parameters are optimized. So why don't use torch.no_grad() in encoder_hidden_states progressing to enhance controlnet & vae_encode speed?

[flymin]

I think we have tried the best to do so. Could you double check? Thanks.

[github-actions[bot]]

This issue is stale because it has been open for 7 days with no activity. If you do not have any follow-ups, the issue will be closed soon.