0

I'm getting a warning after calling load_model in Keras; 

    outfile = self.output_directory+self.data_set_name+'_'+self._model_name()
    model_path = outfile + '_best_model.hdf5'
    model = keras.models.load_model(model_path)

    WARNING:root:The given value for groups will be overwritten.
    WARNING:root:The given value for groups will be overwritten.
    WARNING:root:The given value for groups will be overwritten.
    WARNING:root:The given value for groups will be overwritten.

Do you have any idea what is about?

Thanks

Omran
  • 1
  • I'm not sure whether that warning is from Keras. Could you maybe provide us with a [minimal reproducible example](https://stackoverflow.com/help/minimal-reproducible-example)? – Anakhand Jun 10 '20 at 14:51

1 Answers1

0

If you use Instancenormalization from TF/addons there is your warning in the init call of the instancenorm layer in line 338.

if "groups" in kwargs:
        logging.warning("The given value for groups will be overwritten.")

https://github.com/tensorflow/addons/blob/master/tensorflow_addons/layers/normalizations.py

nEsq
  • 1
  • 2
    When posting links to code on GitHub, press `y` on GitHub to get a permanent link to the code's current revision, so the link doesn't break if the file is moved. Then you can also mark a region of line numbers with shift-click that will stay correct if the file changes. See [here](https://stackoverflow.com/a/31282559/2748899) for details. – smheidrich Mar 09 '21 at 13:14