Help protect the Great Barrier Reef with TensorFlow on Kaggle Join Challenge


Exposes custom classes/functions to Keras deserialization internals.

Used in the notebooks

Used in the guide

Under a scope with custom_object_scope(objects_dict), Keras methods such as tf.keras.models.load_model or tf.keras.models.model_from_config will be able to deserialize any custom object referenced by a saved config (e.g. a custom layer or metric).


Consider a custom regularizer my_regularizer:

layer = Dense(3, kernel_regularizer=my_regularizer)
config = layer.get_config()  # Config contains a reference to `my_regularizer`
# Later:
with custom_object_scope({'my_regularizer': my_regularizer}):
  layer = Dense.from_config(config)

*args Dictionary or dictionaries of {name: object} pairs.



View source


View source