predict

TimeSeriesFoundationModel.predict(data: str | Path | DataFrame, *, target: str = 'target', id_column: str = 'item_id', timestamp_column: str = 'timestamp', known_covariates: str | Path | DataFrame | None = None, static_features: str | Path | DataFrame | None = None, prediction_length: int = 1, quantile_levels: list[float] | None = None, predictions_path: str | None = None, hyperparameters: dict[str, Any] | None = None, instance_type: str | None = None, framework_version: str = '1.6', wait: bool = True, backend_overrides: dict[str, dict[str, Any]] | None = None, **kwargs: Unpack[TrainingJobKwargs]) → DataFrame | JobPredictionFuture[source]

Forecast the future values of data in a one-off SageMaker job.

Parameters:

  • data (str | 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.

  • target (str, default = "target") – Name of the column with the values to forecast.

  • id_column (str, default = "item_id") – Name of the column with the ID of each time series.

  • timestamp_column (str, default = "timestamp") – Name of the column with the observation timestamps.

  • known_covariates (str | Path | pd.DataFrame | None, default = None) – Future values of the known covariates over the forecast horizon. All columns except id_column and timestamp_column are used as known covariates.

  • static_features (str | Path | pd.DataFrame | None, default = None) – Static (time-independent) features describing each individual time series.

  • prediction_length (int, default = 1) – Number of time steps to forecast.

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

  • predictions_path (str | None, default = None) – S3 URL ending in .csv or .parquet where the job writes the predictions, e.g. s3://my-bucket/predictions.csv. The SageMaker execution role needs s3:PutObject permission for it. Defaults to {cloud_output_path}/{job_name}/predictions.csv. The predictions always use the column names item_id and timestamp, regardless of id_column and timestamp_column.

  • hyperparameters (dict[str, Any] | None, default = None) – Model hyperparameters for inference. Overrides values passed to the constructor.

  • instance_type (str | None, default = None) – Instance type for the prediction job. Defaults to the model registry value.

  • framework_version (str, optional) – AutoGluon version, e.g. “1.6”. Uses the official AutoGluon DLC image for this version.

  • wait (bool, default = True) – If True, block until the job completes and return the forecasts. If False, return a JobPredictionFuture immediately; call its .status() to check progress and .result() to get the forecasts.

Returns:

pd.DataFrame | JobPredictionFuture – Forecasts with item_id and timestamp columns, a mean column, and one column per quantile level if wait=True; a JobPredictionFuture otherwise.

Other Parameters:

  • backend_overrides (dict[str, dict[str, Any]] | None, default = None) – Raw SageMaker request fields for settings without a dedicated argument.

    • Keys: request names from the SageMaker API section below.

    • Values: request fields in PascalCase, as in the SageMaker API and boto3. Deep-merged over the request built by AutoGluon-Cloud; lists and other non-dict values replace the generated ones.

    • Example: {"CreateTrainingJob": {"RetryStrategy": {"MaximumRetryAttempts": 2}}}

  • job_name (str | None, default = None) – Name of the training job that runs the prediction. Auto-generated if not set.

  • volume_size (int, default = 100) – Size in GB of the storage volume to use for the job.

  • custom_image_uri (str | None, default = None) – Custom container image URI. If set, framework_version is ignored.

  • timeout (int, default = 86400) – Maximum job runtime in seconds. Defaults to 24 hours.

SageMaker API

  • CreateTrainingJob: runs the prediction as a training job (not a batch transform job) on instance_type. Predictions are written to predictions_path.