Features/region proposal network losses small fixes

broadinstitute / keras-rcnn

Keras package for region-based convolutional neural networks (RCNNs)

Other

553 stars 223 forks source link

Features/region proposal network losses small fixes #55

Closed hgaiser closed 6 years ago

hgaiser commented 6 years ago

This PR fixes some small issues that caused the model not to compile.

In addition I noticed that all inputs are shaped including their batch dimension. I wasn't completely aware of this and I'm not sure the current implementation handles it correctly (ie. it might assume a shape of (None, 4) while it is actually (None, None, 4)). @jhung0 @0x00b1 are you aware of this?

jhung0 commented 6 years ago

@hgaiser
Yeah, I noticed that issue, and we should be consistent about it...so let's try to make sure batch dimension is included

hgaiser commented 6 years ago

Yeah with these changes I got a successful model summary (not really the same as compiling I guess).

jhung0 commented 6 years ago

Are you able to train?

hgaiser commented 6 years ago

No I didn't try training yet (are the loss layers used already?). I only got a seemingly logical model summary.