bigframes.bigquery.ml.arima_evaluate#

bigframes.bigquery.ml.arima_evaluate(model: BaseEstimator | str | Series, *, show_all_candidate_models: bool | None = None) → DataFrame[source]#

Evaluates the model metrics of ARIMA_PLUS or ARIMA_PLUS_XREG time series models.

See the BigQuery ML ARIMA_EVALUATE function syntax for additional reference.

Parameters:

  • model (bigframes.ml.base.BaseEstimator, str, or pd.Series) – The time series model to evaluate.

  • show_all_candidate_models (bool, optional) – A BOOL value that specifies whether to return the evaluation metrics of all the candidate models that auto.ARIMA evaluated, or only those of the best model, which is the one with the lowest Akaike information criterion (AIC). For single time series ARIMA_PLUS or ARIMA_PLUS_XREG models, the default value is True. For large-scale time series ARIMA_PLUS models, the default value is False.

Returns:

The evaluation metrics, one row per model. A model that couldn’t be fitted has NULL metrics and an error_message explaining why.

Return type:

bigframes.pandas.DataFrame