Mahdi Pourmirzaei
Mahdi Pourmirzaei
The value of gaze are about in the range of -3 to 3 and the paper of dataset doesn't describe the normalizing procedure precisely or maybe I couldn't find it....
@VeruGromnicova Thanks for your respond. In the paper said that the head pose angles are between -80 to 80 degree, but if I the labels are in radian, then the...
I think there is a problem if we consider radian for labels. I mean if maximum value of head pose is 180 degree or something like that, then the camera...
Any updates?
I have the same issue with subcellular localization dataset. The number of samples in the training, validation and test sets are different compared to the PEER paper. What is the...
@sduzxj I ran your code and got different numbers: 4945, 95, 394
Same here. What is the difference between these images in the image_log: reconstruction_gs-011070_e-000135_b-000000 samples_cfg_scale_9.00_gs-011070_e-000135_b-000000
For the Image logs part, the recontruction images I think is basde on the output of VQ-VAE. Thats why it looks slightly different from the original image. In other words,...
@eliphatfs would you please show your solution (wrapping models together) in a sudo code? I am working on training controlnet + SD modules together.
> I "solved" it by creating one `Accelerator` per model. If you use only one and register the models via `accelerator.prepare(model1, ..., modelN)` at least one of the models is...