Indexes out of range about pytorch_objectdetecttrack HOT 5 OPEN

Second this! I am getting the same issue for both single image detection and video tracking

@cfotache, is there any chance you could elaborate a bit because the whole idea of using custom trained YOLO with your implementation (the best one I found at least) of SORT is very tempting...

from pytorch_objectdetecttrack.

Hey! I think you should try to change the modules from cuuda to normal Float.Tensor and some part of the program so that it is runs on a local machine without GPU settings. If you are using a GPU then it should work just fun without making any changes

from pytorch_objectdetecttrack.

I just recloned again and changed the settings from GPU to CPU and it worked fine on my local computer.

from pytorch_objectdetecttrack.

I just recloned again and changed the settings from GPU to CPU and it worked fine on my local computer.

The problem is that i do have a GPU with cuda compiled and working on other projects, but i still get the same out of index error when i try to run object_tracker.py

Which parts of the code are responsible for CPU-only runs?

Thanks in advance!

from pytorch_objectdetecttrack.

The linear_assignment function is deprecated in 0.21 and will be removed from 0.23, but sklearn.utils.linear_assignment_ can be replaced by scipy.optimize.linear_sum_assignment You can use from scipy.optimize import linear_sum_assignment as linear_assignment.

from pytorch_objectdetecttrack.