Post x amz targetamazonforecast.createpredictor

v1

latestOpenAPI 3.0.0Apache 2.0 License2026-07-1463278347.3 KB

<note> This operation creates a legacy predictor that does not include all the predictor functionalities provided by Amazon Forecast. To create a predictor that is compatible with all aspects of Forecast, use <a>CreateAutoPredictor</a>. </note> Creates an Amazon Forecast predictor. In the request, provide a dataset group and either specify an algorithm or let Amazon Forecast choose an algorithm for you using AutoML. If you specify an algorithm, you also can override algorithm-specific hyperparameters. Amazon Forecast uses the algorithm to train a predictor using the latest version of the datasets in the specified dataset group. You can then generate a forecast using the <a>CreateForecast</a> operation. To see the evaluation metrics, use the <a>GetAccuracyMetrics</a> operation. You can specify a featurization configuration to fill and aggregate the data fields in the <code>TARGET_TIME_SERIES</code> dataset to improve model training. For more information, see <a>FeaturizationConfig</a>. For RELATED_TIME_SERIES datasets, <code>CreatePredictor</code> verifies that the <code>DataFrequency</code> specified when the dataset was created matches the <code>ForecastFrequency</code>. TARGET_TIME_SERIES datasets don't have this restriction. Amazon Forecast also verifies the delimiter and timestamp format. For more information, see <a>howitworks-datasets-groups</a>. By default, predictors are trained and evaluated at the 0.1 (P10), 0.5 (P50), and 0.9 (P90) quantiles. You can choose custom forecast types to train and evaluate your predictor by setting the <code>ForecastTypes</code>. AutoML If you want Amazon Forecast to evaluate each algorithm and choose the one that minimizes the <code>objective function</code>, set <code>PerformAutoML</code> to <code>true</code>. The <code>objective function</code> is defined as the mean of the weighted losses over the forecast types. By default, these are the p10, p50, and p90 quantile losses. For more information, see <a>EvaluationResult</a>. When AutoML is enabled, the following properties are disallowed: <ul> <li> <code>AlgorithmArn</code> </li> <li> <code>HPOConfig</code> </li> <li> <code>PerformHPO</code> </li> <li> <code>TrainingParameters</code> </li> </ul> To get a list of all of your predictors, use the <a>ListPredictors</a> operation. <note> Before you can use the predictor to create a forecast, the <code>Status</code> of the predictor must be <code>ACTIVE</code>, signifying that training has completed. To get the status, use the <a>DescribePredictor</a> operation. </note>

post/#X-Amz-Target=AmazonForecast.CreatePredictor

Headers

X-Amz-Target'AmazonForecast.CreatePredictor' required

Request body

PredictorNamestring required

A name for the predictor.

AlgorithmArnstring

The Amazon Resource Name (ARN) of the algorithm to use for model training. Required if <code>PerformAutoML</code> is not set to <code>true</code>. Supported algorithms: <ul> <li> <code>arn:aws:forecast:::algorithm/ARIMA</code> </li> <li> <code>arn:aws:forecast:::algorithm/CNN-QR</code> </li> <li> <code>arn:aws:forecast:::algorithm/Deep_AR_Plus</code> </li> <li> <code>arn:aws:forecast:::algorithm/ETS</code> </li> <li> <code>arn:aws:forecast:::algorithm/NPTS</code> </li> <li> <code>arn:aws:forecast:::algorithm/Prophet</code> </li> </ul>

ForecastHorizoninteger required

Specifies the number of time-steps that the model is trained to predict. The forecast horizon is also called the prediction length. For example, if you configure a dataset for daily data collection (using the <code>DataFrequency</code> parameter of the <a>CreateDataset</a> operation) and set the forecast horizon to 10, the model returns predictions for 10 days. The maximum forecast horizon is the lesser of 500 time-steps or 1/3 of the TARGET_TIME_SERIES dataset length.

ForecastTypesForecastType[]

Specifies the forecast types used to train a predictor. You can specify up to five forecast types. Forecast types can be quantiles from 0.01 to 0.99, by increments of 0.01 or higher. You can also specify the mean forecast with <code>mean</code>. The default value is <code>["0.10", "0.50", "0.9"]</code>.

PerformAutoMLboolean

Whether to perform AutoML. When Amazon Forecast performs AutoML, it evaluates the algorithms it provides and chooses the best algorithm and configuration for your training dataset. The default value is <code>false</code>. In this case, you are required to specify an algorithm. Set <code>PerformAutoML</code> to <code>true</code> to have Amazon Forecast perform AutoML. This is a good option if you aren't sure which algorithm is suitable for your training data. In this case, <code>PerformHPO</code> must be false.

AutoMLOverrideStrategy'LatencyOptimized' | 'AccuracyOptimized'

<note> The <code>LatencyOptimized</code> AutoML override strategy is only available in private beta. Contact Amazon Web Services Support or your account manager to learn more about access privileges. </note> Used to overide the default AutoML strategy, which is to optimize predictor accuracy. To apply an AutoML strategy that minimizes training time, use <code>LatencyOptimized</code>. This parameter is only valid for predictors trained using AutoML.

PerformHPOboolean

Whether to perform hyperparameter optimization (HPO). HPO finds optimal hyperparameter values for your training data. The process of performing HPO is known as running a hyperparameter tuning job. The default value is <code>false</code>. In this case, Amazon Forecast uses default hyperparameter values from the chosen algorithm. To override the default values, set <code>PerformHPO</code> to <code>true</code> and, optionally, supply the <a>HyperParameterTuningJobConfig</a> object. The tuning job specifies a metric to optimize, which hyperparameters participate in tuning, and the valid range for each tunable hyperparameter. In this case, you are required to specify an algorithm and <code>PerformAutoML</code> must be false. The following algorithms support HPO: <ul> <li> DeepAR+ </li> <li> CNN-QR </li> </ul>

TrainingParametersobject

The hyperparameters to override for model training. The hyperparameters that you can override are listed in the individual algorithms. For the list of supported algorithms, see <a>aws-forecast-choosing-recipes</a>.

OptimizationMetric'WAPE' | 'RMSE' | 'AverageWeightedQuantileLoss' | 'MASE' | 'MAPE'

The accuracy metric used to optimize the predictor.

Response

Success

PredictorArnstring

The Amazon Resource Name (ARN) of the predictor.