NaN loss when object has category not in dictionary

[jhung0]

If the dictionary of categories does not contain a category that an object has, the loss becomes NaN

[0x00b1]

An exception should be thrown (or warning should be returned) if the generator encounters a category that isn’t a member of the categories dictionary. What do you think?

[jhung0]

Ideally a warning and make it just skip that object

[0x00b1]

@jhung0 I looked into this yesterday and couldn't reproduce. When I added a non-existent category the ObjectDetectionGenerator threw a missing key exception. Would you mind posting an example?

[jhung0]

training_dictionary, test_dictionary = keras_rcnn.datasets.shape.load_data()

categories = {"circle": 1, "rectangle": 2}