New answers tagged

2 votes

In order to predict after training on Standardized data, do I need the StandardScaler as well?

You need to save and restore the scaler independently of the keras model - the scaler is from scikit-learn. This is easier done than said: ...
brewmaster321's user avatar
0 votes

My custom neural network is converging but keras model not

In this instance, the default learning rate for the adam optimizer is 0.001 in keras, so it will take a very very long time to converge. If you bump this up as follows: ...
brewmaster321's user avatar
1 vote

What exactly is saved when I save a NN?

The saved data comprises: Model's architecture: so, which layers your model has (Masking, Dense, Dense, Dense), their definition (e.g. the input/output shapes, activations, ...
noe's user avatar
  • 25.7k

Top 50 recent answers are included