ayushexel / detectx-yolo-v3 Goto Github PK

Hello,

I've run into a problem with the CuDNN (specifically, CuDNN_STATUS_INTERNAL_ERROR).

Therefore, until I can get CuDNN to work, is there a way to run the DetectX script without Cuda?

Like having an else case for the following if and similar ones?

if CUDA:
            batch = batch.cuda()

When I try to run the model on the video I see this error.

Traceback (most recent call last):
  File "detect_video.py", line 110, in <module>
    img, orig_im, dim = prepare_input(frame, inp_dim)
  File "detect_video.py", line 26, in prepare_input
    img = (resize_img(orig_im, (inp_dim, inp_dim)))
NameError: name 'resize_img' is not defined

Please let me know where i am going wrong...........
Please help me asap..