After you train a model in Machine Learning Designer, you can generate predictions on new data by using batch predictions or real-time predictions, depending on your timeliness requirements.
-
Batch predictions
Add the trained model and test data to prediction components in Machine Learning Designer for batch prediction. You can then submit the pipeline to DataWorks and schedule it as a periodic task. For more information, see Batch predictions.
-
Real-time predictions
-
Deploy a single model as an online service
Deploy models as online services in Elastic Algorithm Service (EAS) for real-time prediction. Push-button deployment is available for Predictive Model Markup Language (PMML), AlinkModel, and XGBoost models trained by Machine Learning Designer. You can also manually export PMML model files to import in EAS. Models in the Parameter Server (PS) format require manual export before you can deploy them as EAS online services.
-
Deploy a pipeline as an online service
Deploy pipelines as online services for real-time prediction. Use Alink algorithm components to build a batch data processing pipeline that handles data preprocessing, feature engineering, and model prediction. Package the pipeline as a pipeline model and deploy it as an EAS online service with a few clicks.
-