PCT issues

The question of training

I'm very sorry to bother you. I trained the PCT in a 2080Ti，but I can‘t obtain the accuracy to 93.2 and the result is greatly waving. I only obtain the...

Hello-hubKing

Are there potential errors in the implementation of SA?

4

Hi. As [here](https://github.com/MenghaoGuo/PCT/blob/main/networks/cls/pct.py#L221), the attention matrix should be transposed before the matrix product, if I understand it correctly. Here is my draft of the calculation about the dimension of the...

jialeli1

partseg and semseg

hi.I had some problems reproducing your paper.Can you release the complete code?The partseg and semseg I reproduced with pytorch are bad.

mmiku1

I want to know how you visualize your attention map

11

Hi, First of all, I want to thank you for the proposed method , which benefited me a lot. So I reproduce your code by pytorch and tried to visualize...

ja604041062

How do you consider about Layer normalization and Batch normalization?

Hi, PCT used the Batch normalization, instead of the Layer normalization, used by original Transformer. I wonder how do you consider about Layer normalization and Batch normalization in PCT?

Alobal

What is `self.pos_xyz(xyz)`

2

Hello, thanks for your work! I am reading the code but get confused about some details. In the `class Point_Transformer_Last`, the `self.pos_xyz` is not defined in the section. What does...

songtianhui

About SA_Layer

2

When the convolution weight of q and k are initialized equal, `self.q_conv.conv.weight = self.k_conv.conv.weight` will it cause q and k to be always the same when updated?

Hzj199

Questions about segmentation

4

Thanks for your sharing! I benefit a lot from the code. However, I also meet some problems. Based on the released segmentation code, I conducted the experiment on the S3DIS...

accoutzzz

The performance of segmentation

5

Thank you for sharing model code selflessly! Recently I have reproduced the segmentation based on provided seg code, however the class mean IoU is only 78.8%. Could you please offer...

tegusi