胡钰程

> out_selected_a = Variable(torch.from_numpy(out_a.cpu().data.numpy()[hard_triplets]).cuda()) > out_selected_p = Variable(torch.from_numpy(out_p.cpu().data.numpy()[hard_triplets]).cuda()) > out_selected_n = Variable(torch.from_numpy(out_n.cpu().data.numpy()[hard_triplets]).cuda()) will lead to > RuntimeError: element 0 of tensors does not require grad and does not have a...

save problem

5

how can I save the output as png or jpg?

I want to know the detail or source code about the pre-processed of human 3.6M

No permission，can not download the dataset

https://github.com/OpenDriveLab/ViDAR/blob/936dbf7e010189b68b83b4b61568cfd0fa23e655/projects/mmdet3d_plugin/bevformer/dense_heads/vidar_head_base.py#L736 请问这里通过scale_factor实现grid坐标到bev坐标的转换，但z和x、y的缩放比例是不一样的，会导致decode_dict里的gt_pcds不等同于输入的gt_points，这样测试不会有什么问题嘛