Open ErenKaymakci opened 1 year ago
Hi, did you find a solution?
What is the input image size of your YOLOv4 network (from the .cfg file)?
If it's different than 416, you will need to specify that with the run command. Such as:
python save_model.py --model yolov4 --input_size 512
I changed .cfg file for small objects like here: https://github.com/AlexeyAB/darknet#how-to-improve-object-detection and when I run
python save_model.py --model yolov4
, I got this error: File "/content/yolov4-deepsort/core/utils.py", line 63, in load_weights conv_weights = conv_weights.reshape(conv_shape).transpose([2, 3, 1, 0]) ValueError: cannot reshape array of size 8192 into shape (24,1024,1,1)I suggest updating the repo as we can add the cfg file path and the algorithm gets all the parameters there.