3.1.3 HeatWave AutoML Workflow

A typical HeatWave AutoML workflow is described below:

1. When the ML_TRAIN routine is called, HeatWave AutoML calls the MySQL DB System where the training data resides. The training data is sent from the MySQL DB System and distributed across the HeatWave Cluster, which performs machine learning computation in parallel. See Section 3.5, “Training a Model”.

2. HeatWave AutoML analyzes the training data, trains an optimized machine learning model, and stores the model in a model catalog on the MySQL DB System. See Section 3.13.1, “The Model Catalog”.

3. HeatWave AutoML ML_PREDICT_* and ML_EXPLAIN_* routines use the trained model to generate predictions and explanations on test or unseen data. See Section 3.7, “Predictions”, and Section 3.8, “Explanations”.

4. Predictions and explanations are returned to the DB System and to the user or application that issued the query.

Optionally, the ML_SCORE routine can be used to compute the quality of a model to ensure that predictions and explanations are reliable. See Section 3.13.6, “Scoring Models”.

Note

HeatWave AutoML shares resources with HeatWave. HeatWave analytics queries are given priority over HeatWave AutoML queries. Concurrent HeatWave analytics and HeatWave AutoML queries are not supported. A HeatWave AutoML query must wait for HeatWave analytics queries to finish, and vice versa.