PrincipalComponentsModel train

---

train(self, frame, observation_columns, mean_centered=True, k=None)

Build principal components model.

Parameters:

frame : Frame

A frame to train the model on.

observation_columns : list

List of column(s) containing the observations.

mean_centered : bool (default=True)

Option to mean center the columns

k : int32 (default=None)

Principal component count. Default is the number of observation columns

Returns:

: dict

dictionary

|A dictionary with trained Principal Components Model with the following keys: |‘column_means’: the list of the means of each observation column |‘k’: number of principal components used to train the model |‘mean_centered’: Flag indicating if the model was mean centered during training |‘observation_columns’: the list of observation columns on which the model was trained, |‘right_singular_vectors’: list of a list storing the right singular vectors of the specified columns of the input frame |‘singular_values’: list storing the singular values of the specified columns of the input frame

Creating a PrincipalComponents Model using the observation columns.

Examples

See here for examples.