JefferyChiang comments

Results 8 comments of


                                            JefferyChiang

Can this model deal with the dataset with different size images?

The input size of ShuffleNetV2 from pytorch is 224*224. I wonder how does it work when input size is set to 320 or 416. Will NanoDet resize img to 224...

Detect usages of torch.jit and collect its required source files automatically

Thanks This solution still work for pyinstaller==4.5.1. and timm==0.5.4.

Streaming video update

#1475

module 'sympy.core.core' has no attribute 'symbol'

upgrade sympy to 1.10.1 solve the problem

A new `forward_distmat` method to accelerate distance matrix computing among images.

px.imshow(facet_col_wrap) messes with facet_col index

Same issue here. Do you have any solution for this ?

Thanks for your work. But it's a bit strange here in the coreset sampling part

I agree with your point. I think the first if-branch should be modify into "if not self.already_selected: " for more accurate to the kcenter_greedy algorithm. Have you modify the code...

Code on computing the anomaly score is inconsistent with the paper

It can be changed to "w = (1 - (np.exp(**N_b[0]**)/np.sum(np.exp(N_b))))" because the N_b is sorted by k. In the other hand, I want to figure out some explanation of why...