Where are you putting this loss.py function? Also I am trying to use this to train on my own data but I keep getting empty feature map. Do you know how to train on my own data?

CasiaFan / tensorflow_retinanet

RetinaNet with Focal Loss implemented by Tensorflow

121 stars 45 forks source link

Where are you putting this loss.py function? Also I am trying to use this to train on my own data but I keep getting empty feature map. Do you know how to train on my own data? #17

Open ishacusp opened 5 years ago

CasiaFan commented 5 years ago

@ishacusp TF object detection has its own focal loss function integrated into its framework and you could enable it by defining weighted_sigmoid_focal in config loss section. If you'd like to use this customed loss function, modify object_detection/meta_architecture/ssd_meta_arch to replace the default classification loss (Line 844).

ishacusp commented 5 years ago

Thanks! I followed your README and used your retinanet.py and feature extractor, to train on my data, but the feature extractor returns an empty map.

CasiaFan commented 5 years ago

@ishacusp Okay... I get it. This error is caused by inexplicit parameter assignment during retinanet feature extractor initialization when I add min_level and max_level in base class only. I have fixed this bug and you could pull it down to use. Thanks for reporting.

ishacusp commented 5 years ago

Thank You!