JYounngS

Results 3 comments of JYounngS

Why BEVDepthOCC gets worse results than BEVDetOCC ?

Has anyone tried to use the BEVDepth as your baseline?

RuntimeError: The expanded size of the tensor (3) must match the existing size (4) at non-singleton dimension 3. Target sizes: [1, 1, 3, 3]. Tensor sizes: [4, 4]

> I have encountered the same problem, have you solved it? I rolled back the changes in issue 332 and triggered a new error > > ```python > # issue...

RuntimeError: The expanded size of the tensor (3) must match the existing size (4) at non-singleton dimension 3. Target sizes: [1, 1, 3, 3]. Tensor sizes: [4, 4]

> 报错位置代码： bda_[:, :, :3, :3] = bda.unsqueeze(1) > > 修改为：bda_[:, :, :, :] = bda.unsqueeze(1) 即可。 Can you get the right mAP and NDS after this modification?