Home > Software design >  Is there a way to use a data generator in a KerasClassifier function?
Is there a way to use a data generator in a KerasClassifier function?

Time:01-18

According to this doc file of TensorFlow, the input data may accept a data generator. However, the KerasClassifier explicitly accepts x and y attributes. How can I go around this since I would want to use a data generator for KerasClassifier?

When I am ingesting a data generator on KerasClassifier, I am getting an error that goes along the lines of "KerasClassifier is missing a y attribute".

CodePudding user response:

The answer is probably no.

The KerasClassifier class is meant to be used with scikit-learn APIs, which do not have a concept of a data generator.

So it has not be implemented, because there is no API to implement. If you want to train your model like this, you should use the Keras fit API directly.

  • Related