predict

TimeSeriesEndpoint.predict(data: str | Path | DataFrame, *, known_covariates: str | Path | DataFrame | None = None, static_features: str | Path | DataFrame | None = None, prediction_length: int | None = None, target: str | None = None, id_column: str | None = None, timestamp_column: str | None = None, quantile_levels: list[float] | None = None) → DataFrame[source]

Forecast the future values of data with the deployed endpoint.

On trained predictor endpoints, prediction_length, target, id_column, timestamp_column, and quantile_levels default to the values set at fit time, and setting prediction_length, target, or quantile_levels to a different value raises an error.

Parameters:

  • data (str | pathlib.Path | pd.DataFrame) – Historical time series to forecast from, in long format, as a pd.DataFrame or local/S3 path to a data file. See the TimeSeriesPredictor docs for the expected format.

  • known_covariates (str | pathlib.Path | pd.DataFrame | None, default = None) – Future values of the known covariates over the forecast horizon, as a pd.DataFrame or local/S3 path to a data file.

  • static_features (str | pathlib.Path | pd.DataFrame | None, default = None) – Static (time-independent) features describing each individual time series, as a pd.DataFrame or local/S3 path to a data file.

  • prediction_length (int | None, default = None) – Number of time steps to forecast. Defaults to 1 on foundation model endpoints.

  • target (str | None, default = None) – Name of the column with the values to forecast. Defaults to "target" on foundation model endpoints.

  • id_column (str | None, default = None) – Name of the column with the ID of each time series. Defaults to "item_id" on foundation model endpoints.

  • timestamp_column (str | None, default = None) – Name of the column with the observation timestamps. Defaults to "timestamp" on foundation model endpoints.

  • quantile_levels (list[float] | None, default = None) – Quantiles to forecast, as floats between 0 and 1. Defaults to [0.1, 0.2, ..., 0.9] on foundation model endpoints.

Returns:

pd.DataFrame – Forecasts with item_id and timestamp columns, a mean column, and one column per quantile level.

SageMaker API

  • InvokeEndpoint: sends the data to the endpoint and returns the predictions. The payload is limited to 6 MB (4 MB for serverless endpoints).