faster_rcnn_pytorch icon indicating copy to clipboard operation
faster_rcnn_pytorch copied to clipboard

Published 20 hours ago • verified publisher icon longcw

The training process is not the same as the paper.

Open mariolew opened this issue 8 years ago • 1 comments

The paper use as training method called "Alternating Training", but the code just train the model end to end.

mariolew avatar Jun 14 '17 05:06 mariolew

Paper also talks about end2end, its 1.5x faster at only slight decrease in accuracy.

Rizhiy avatar Jul 19 '17 12:07 Rizhiy