Scaling Data

[MarleneBs]

I have manually scaled a dataframe including features in GLuonTS with a scaling rule I selected. Then I ran the training and forecasting process. Now I have a forecast_object from the package gluonts.model.forecast module and a "test object" where the comparison values are. But now I have unnormalise this values again in order to draw the forecast and its evaluation. I wonder if anyone knows how to do this, i.e. how to manually unnormalise these two objects again before I produce error metrics and a drawing with the corresponding GluonTS functions.

[Msaleh87]

In your estimator, you could have could enable scaling by: scaling = True
You don't have to do this manually.

[MarleneBs]

Thank you, but this only works for the target value in deepar, does'nt it? For the features there is not possibility?