search

hhk7734 / tensorflow-yolov4

YOLOv4 Implemented in Tensorflow 2.
MIT License
136 stars 75 forks source link

What is the purpose of Dataset.grid in py_src/yolov4/tf/dataset.py #44

Closed daniel-SX closed 3 years ago

daniel-SX commented 3 years ago

In the Dataset object you defined a grid attribute which is only used in line 182: https://github.com/hhk7734/tensorflow-yolov4/blob/fd45eea11d2de86664978184e6b798daf2546cdc/py_src/yolov4/tf/dataset.py#L182-L183

I don't understand the purpose of this assignment , because later the xywh gt vector is assigned to each grid point in line 222: https://github.com/hhk7734/tensorflow-yolov4/blob/fd45eea11d2de86664978184e6b798daf2546cdc/py_src/yolov4/tf/dataset.py#L222

And when calculating the xiou_loss only these xywh gt vectors are used. So there seems to be no use anywhere for the _grid values.

hhk7734 commented 3 years ago

The grid's x and y values ​​are pre-assigned in case they are used in calculations. If it is unnecessary and has an impact on the training time, I think we can erase it after testing. :)