<model_name>!EXPLAIN_FEATURE_IMPORTANCE

Returns the relative feature importance for each feature used by the model.

If you need to select specific columns from the data returned by this method, you can call the method in the FROM clause of a SELECT statement. See Selecting columns from SQL class instance methods that return tabular data.

Syntax

<model_name>!EXPLAIN_FEATURE_IMPORTANCE();

Returns

ColumnTypeDescription
SERIESVARIANTSeries value (NULL if model was trained with single time series).
RANKINTEGERThe importance rank of a feature for a specific series
FEATURE_NAMEVARCHAR

The name of the feature used to train the model aggregated_endogenous_features represents all features derived as transformations of your target variable.

IMPORTANCE_SCOREFLOATThe feature’s importance score: a value in [0, 1], with 0 being the lowest possible importance, and 1 the highest.
FEATURE_TYPEVARCHAR

The source of the feature, one of:

  • user_provided
  • derived_from_timestamp
  • derived_from_endogenous