Unlock the Power of Hyperparameter Tuning: How AutoML Simplifies HRNN Recipe Selection in Amazon Personalize
- Amazon Personalize offers an automated solution for selecting the most suitable Hierarchical Recurrent Neural Network (HRNN) recipe based on input data analysis.
- To utilize AutoML, set the performAutoML parameter to true when calling the CreateSolution API.
- Alternatively, you can specify a list of recipes for Amazon Personalize to consider and determine the best one based on a specified metric.
Optimizing Amazon Personalize with AutoML
Amazon Personalize offers an automated solution for selecting the most suitable Hierarchical Recurrent Neural Network (HRNN) recipe based on input data analysis. This feature is known as AutoML.
To utilize AutoML, set the performAutoML parameter to true when calling the CreateSolution API. This allows Amazon Personalize to automatically choose the best recipe for your specific use case.
Alternatively, you can specify a list of recipes for Amazon Personalize to consider and determine the best one based on a specified metric. To do this, call the CreateSolution operation with performAutoML set to true and omit the recipeArn parameter. Include the solutionConfig parameter and specify the metricName and recipeList as part of the autoMLConfig object.
Choosing a Recipe with Amazon Personalize
| performAutoML | recipeArn | solutionConfig | Outcome |
|---|---|---|---|
| true | Omit | Omit | Amazon Personalize will choose the recipe for you |
| true | Omit | autoMLConfig: Specify metricName and recipeList | Amazon Personalize will choose a recipe from the list to optimize the metric |
| Omit | Specify | Specify | Choose your own recipe to override the default training properties |
When performAutoML is set to true, all parameters in the solutionConfig object are ignored, except for autoMLConfig.