Closed Rizhiy closed 7 years ago
In particular I'm getting empty overlaps in faster_rcnn/rpn_msr/anchor_target_layer.py:144
.
A lot of my training images don't contain any objects, can this be a problem?
I found a handler in another repo, which seems to work. I needed Caltech pedestrian handler, check my fork for more info.
How should I add new dataset to train on?
I am confused by all the functions in each python file. Which ones do I need to replace?