TypeError: unsupported array index type list(int64)<iv=None> in Tuple(slice<a:b>, list(int64)<iv=None>, slice<a:b>)

pangsu0613 / CLOCs

CLOCs: Camera-LiDAR Object Candidates Fusion for 3D Object Detection

MIT License

351 stars 68 forks source link

TypeError: unsupported array index type list(int64)<iv=None> in Tuple(slice<a:b>, list(int64)<iv=None>, slice<a:b>) #60

Closed SunshineKK456 closed 2 years ago

SunshineKK456 commented 2 years ago

hello,I meet some troubles when I trained the CLOCs:

There are 22 candidate implementations:

Of which 20 did not match due to: Overload of function 'getitem': File: : Line N/A. With argument(s): '(array(float32, 3d, C), Tuple(slice, list(int64), slice))': No match.
Of which 2 did not match due to: Overload in function 'GetItemBuffer.generic': File: numba/core/typing/arraydecl.py: Line 162. With argument(s): '(array(float32, 3d, C), Tuple(slice, list(int64), slice))': Rejected as the implementation raised a specific error: TypeError: unsupported array index type list(int64) in Tuple(slice, list(int64), slice)

When I finished 3700 steps,I meet the above question,Can you give me some help or advice？Thanks.

SunshineKK456 commented 2 years ago

I find some solutions on the Internet,just like change the function predict_v2:

opp_labels=(box_preds[...,-1] >0) ^dir_lables.byte() → opp_labels=(box_preds[...,-1] >0) ^dir_lables.to(torch.bool)

But it doesn't work.....

pangsu0613 commented 2 years ago

Hello @SunshineKK456 , I think I need more information regarding your error. May I know in what situation you encountered these errors? are you training with other 2D or 3D detections? Could you paste all the error message from the terminal? Thank you.

SunshineKK456 commented 2 years ago

@pangsu0613 hello, I have solved the trouble. Thanks for your excellent work !

pangsu0613 commented 2 years ago

@SunshineKK456, that's great! May I know is it related to "opp_labels=(box_preds[...,-1] >0) ^dir_lables.byte() → opp_labels=(box_preds[...,-1] >0) ^dir_lables.to(torch.bool)" in 'predict_v2' ?

SunshineKK456 commented 2 years ago

@pangsu0613 I see a same issue before, and followed the steps. Someone suggest me to change the labels‘ data type. As you can see, I tried to change the 'byte()' into the 'bool' in pred(there are no mistakes in the training), but it doesn't work. By the way, I like saber too.

IsFaqq commented 1 year ago

@SunshineKK456 hello,I had the same problem, how did you end up solving it? I also tried to change the 'byte()' into the 'bool' in pred,but it doesn't work.