luo_13 comments

Results 9 comments of


                                            luo_13

Center of volume

> > Do you think that he was wrong? > > Yes. I think this pytorch version is incomplete compared to the orginal version written in Theano. > In this...

Center of volume

> We don't need to update value of c (center of volume) during the training? Yes, We don't need to update the value of c (center of volume) during the...

exp文件中将self.mosaic_prob置为0.5时报错 'int' object has no attribute 'numel'

Same problem.

exp文件中将self.mosaic_prob置为0.5时报错 'int' object has no attribute 'numel'

[#658](https://github.com/Megvii-BaseDetection/YOLOX/issues/658) solve the problem.

FeatureRequest : Model access control solution

Any progress on this?

Issue about the dataset used in your code

> @seekFire > Thx too! Keypoints[:, 2] represents the mean value of three nearest point distances. > 都是中国人本来想中文回复的，但是为了国外友人能看懂还是英文回复了 > 加油 :) I think this is not the nearest neighbor. I...

document use cases for onSubmittedWorkDone

> I'm still wondering whether the "get a copy of the image contents of a context" algorithm gets executed right away when I create the `VideoFrame`, or whether implementers may...

Why the code for Wan2.1 uses t_mod to determine whether to cache rather than using timestep embedding?

Same question — why do we even need a timestep embedding if it’s only a function of t and freq_dim, and has nothing to do with the model’s actual input...

Why the code for Wan2.1 uses t_mod to determine whether to cache rather than using timestep embedding?

The paper states: "As for the timestep embedding, it changes as timesteps progress but is independent of the noisy input and text embedding, making it difficult to fully reflect the...