Do you understand why the Outputs of my YOLOV4 model has this shape?

Question

I have a question. I trained a YOLOV4 model for face detection and when i tried to look at the output on Neutron i found that the Bounding Box shape is [1,1,1] and Feature has the shape of [1,1,4] is there something wrong that i did ?

Thank you![Neutron] https://i.stack.imgur.com/jnj1h.png

Please provide enough code so others can better understand or reproduce the problem. — KiynL, May 13 '22 at 10:57

score 2 · Answer 1 · answered May 13 '22 at 11:16

2

This is already answered here. But basically it's like this:

(1,1,4) is the result for the bounding boxes. The first number 1 is according to your image fed into the model. As you have only one object then the output is 1 for the second number. Furthermore, YOLO configuration for bounding boxes are (x_center,y_center,width,height), which are the 4 positions.

(1,1,1) will be the same, but now 1 is for the label of the class you choose.

answered May 13 '22 at 11:16

Noltibus

1,300
3
12
34

If there is already an answer in Stack Overflow, there is no point in answering again. – Dr. Snoopy May 13 '22 at 11:20
It has been done thousands of times... – Noltibus May 13 '22 at 11:21
And? That is irrelevant. – Dr. Snoopy May 13 '22 at 11:23
@Dr.Snoopy i said this because i have this error that i don't understand, so if you can help : java.lang.IllegalArgumentException: Size of bounding box dimension 2 is not 4. Got 1 in shape [1, 2535, 1] – Reda Ctl May 17 '22 at 15:17

Do you understand why the Outputs of my YOLOV4 model has this shape?

1 Answers1