search

jwyang / faster-rcnn.pytorch

A faster pytorch implementation of faster r-cnn
MIT License
7.71k stars 2.33k forks source link

Why gt_boxes all 1 in testing? #874

Open sjytker opened 2 years ago

sjytker commented 2 years ago

I run test_net.py and got all 0 results. I have checked all the related issue and nothing could be done. I notice that in roibatchLoader.py, gt_boxes are make all 1 in this code gt_boxes = torch.FloatTensor([1, 1, 1, 1, 1]) in line 210 or so. Is it correct?

xlhuang132 commented 2 years ago

I run test_net.py and got all 0 results. I have checked all the related issue and nothing could be done. I notice that in roibatchLoader.py, gt_boxes are make all 1 in this code gt_boxes = torch.FloatTensor([1, 1, 1, 1, 1]) in line 210 or so. Is it correct?

Do you know why? I'm also wondering

sjytker commented 2 years ago

I run test_net.py and got all 0 results. I have checked all the related issue and nothing could be done. I notice that in roibatchLoader.py, gt_boxes are make all 1 in this code gt_boxes = torch.FloatTensor([1, 1, 1, 1, 1]) in line 210 or so. Is it correct?

Do you know why? I'm also wondering

I don't know, maybe the author abandon the repo in this way…… I have finished my job using another repo.