ShenXiaolei

Is it divided into 256 buckets based on the amplitude of the optical flow (linear or log)?

overfit phenomenon?

1

Thanks for your great work. Have you ever encounter the phenomenon of overfit?

ReferenceAttentionControl

1

With unconditional generation during training, should reference embedding concat to the normal_hidden_states?

variable aspect ratios, resolutions, durations

3

Is the implementation of variable aspect ratios, resolutions, durations different from that of NaViT? Are there any plans to implement the NaViT?

MobileNet-V3 checkpoints

1

Tnanks for the great work. Could you please provide the checkpoints of the MobileNet V3 model? Thanks!

pipeline at dev branch

7

at dev brach, init pipeline as following code, but the output image is covered with a red layer `# brushnet-based version unet = UNet2DConditionModel.from_pretrained( "stable-diffusion-v1-5/stable-diffusion-v1-5", subfolder="unet", revision=None, torch_dtype=weight_dtype, ) text_encoder...

detach

1

why use detach() for promptB？Is it the solution for "Parameter at index 1 with name xxx has been marked as ready twice"?