czhxiaohuihui

I am wondering how to visualize attention as Figure 6 in your paper? Can you share code with this?

About extract visual relation embedding

1

Can you explain how to use Large-Scale-VRD to extract visual relation embedding exactly？Very Thanks!!!!

how to process the RPLAN dataset

1

I use the code in misc/read_floorplan.py to process the RPLAN dataset, but I find the bboxes is not computed，and always [] !!!!! So, how to get the bboxes bboxes, edges,...

蒸馏的效果没有直接用数据训练TextCNN的效果好？

2

我在自己的数据集上试了一下，bert的效果大概是85%，textCNN是79%， 然后用蒸馏大概只有77.8%. 蒸馏相关的两个参数都是按照你代码里的： self.T = 10 # 调整温度 self.alpha = 0.9 # 调整soft_target loss 和 hard_target loss 比重