Faster RCNN unbalanced dataset


I would like to train a Faster RCNN model with an unbalances dataset

  • class A : 1800 objects / bounding boxes
  • class B: 150 objects / bounding boxes

Is there some parameters to improve the training ?

For faster_rcnn, currently there are not relevant parameters.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.