[Question] Extract training data and target from a model at timestep t

[nkonts]

Hi,

I am currently exploring your library and working through some example notebooks. To gain more confidence in what is happening under the hood, is there a possiblity to extract the X and y similar to what is shown as resulting arrays in the Lagged data extraction image from the Regression Models example?

I was not able to find a way while going through your API reference or inspecting the attributes of the LinearRegressionModel object.

[madtoinou]

Hi @nkonts,

The RegressionModel._create_lagged_data() method is responsible for creating the X and y arrays. Under the hood, it calls create_lagged_training_data() from the tabularization module (doc).

I hope that it solves your problem.