myt889

Need the code for the multi-label dataset Pascal VOC

5

Great work!. Can you provide the code for the multi-label dataset Pascal VOC for reproducing the result? Wait for your reply. Best regards!

能否提供EPSANet-50在ImageNet上的预训练权重？

8

非常感谢！

Several Questions

2

Hi, Thank you for your great work, I am new to this area, I have the following questions: 1、How much improvements can be made by the branch of Graph-based contrastive...

Code for visualization of multi-head attention maps

1

Hi Liu, Thank for your great work and sharing clean code. Can you share your visualization code of multi-head attention maps，I am new to transformer, and find it difficult to...

Hi, thank you for your great work, I would like to know if AdaFace can be used for classification task? If yes, how can I adapt it to classification task?...

您好，感谢您的代码，有几个问题想要请教您： 目的：可视化Query2Label这个多标签分类模型的decoder中的multi-head attention. 1. 首先Query2Label模型是由CNN backbone+ Transformer decoder组成的，其中decoder部分由torch.nn .MultiheadAttention()构成，比如有4个heads。CNN backbone，Transformer网络分别在单独的.py文件搭建。 目前我已经在Transformer.py文件中按您的建议步骤，在指定位置输入代码： from visualizer import get_local @get_local("attention_map") 并且在需要网络输出attention的def forward()前，输入代码： @get_local('attention_map') def forward(self, tgt,....): .......... attention_map = out return out 2. 按您的指南，写了对应的可视化代码。...