ValueError: Model and weights file do not match.

[lidorshimoni]

I'm suddenly getting this error... but it worked for me before with the same weights file. using yolov4-tiny

ValueError: Model and weights file do not match.

[lidorshimoni]

even with new and unchanged yolov4.weights and coco.names files, it raises the same exception.

[hhk7734]

Can you share your script?

[Gabrielopesantos]

I'm getting the same error with yolov4 and yolov4-tiny. I'm using the "tutorial" script present on the documentation page.

[hhk7734]

from yolov4.tf import YOLOv4

yolo = YOLOv4()
# yolo = YOLOv4(tiny=True)

yolo.classes = "coco.names"
yolo.input_size = (640, 480)

yolo.make_model()
yolo.load_weights("yolov4.weights", weights_type="yolo")
# yolo.load_weights("yolov4-tiny.weights", weights_type="yolo")

yolo.inference(media_path="kite.jpg")

yolo.inference(media_path="road.mp4", is_image=False)

is it?

[Gabrielopesantos]

Yes. I did a little investigation and at some point the length of the data read from the weights file is different from the count (darknet shape product).

[hhk7734]

Check version.

python3 -m pip show yolov4

Name: yolov4
Version: 2.0.0
...

And try this.

from yolov4.tf import YOLOv4

yolo = YOLOv4(tiny=True)
yolo.classes = "coco.names"
assert(len(yolo.classes) == 80)
yolo.input_size = (640, 480)

yolo.make_model()
yolo.load_weights("yolov4-tiny.weights", weights_type="yolo")
yolo.inference("kite.jpg")

[Gabrielopesantos]

Problem solved, thanks. The error was in my coco.names file.

[hhk7734]

@lidorshimoni Closed because there is no answer for a long time.