Hello
Could I do the same thing with accuracy?

Hello. Thank you for your amazing work with GUI. I'de like to use with this GUI my own weights from own dataset. Could you explain if I need to change code -> downloading from darknet (#56 net = cv2.dnn.readNetFromDarknet(configPath, weightsPath)) or I should only change the paths to my own classes and weights files/directories?
If I run "python yolo_video.py" I get the following error:

[INFO] loading YOLO from disk...
Traceback (most recent call last):
File "yolo_video.py", line 56, in
net = cv2.dnn.readNetFromDarknet(configPath, weightsPath)
cv2.error: OpenCV(4.1.2) C:\projects\opencv-python\opencv\modules\dnn\src\darknet\darknet_importer.cpp:214: error: (-212:Parsing error) Failed to parse NetParameter file: yolo-coco\yolov3.weights in function 'cv::dnn::dnn4_v20190902::readNetFromDarknet'

if I try to change the original weights from YOLO to my own as .h5 then I get:

[INFO] loading YOLO from disk...
Traceback (most recent call last):
File "yolo.py", line 65, in
net = cv2.dnn.readNet(configPath, weightsPath)
cv2.error: OpenCV(4.2.0) C:\projects\opencv-python\opencv\modules\dnn\src\darknet\darknet_io.cpp:712: error: (-213:The function/feature is not implemented) Transpose the weights (except for convolutional) is not implemented in function 'cv::dnn::darknet::ReadDarknetFromWeightsStream'

Thank you for your help!

Hello!
Please tell me how to run this on GPU. I'm not able to configure GPU access. This code doesn't run on GPU (runs on CPU with low frame-rate). Thanks in advance.

https://www.dropbox.com/s/0pq7le6fwtbarkc/yolov3.weights?dl=1