Once you have access to a DB System with a HeatWave Cluster, and you have obtained the MySQL user privileges described in Section 3.1, “Before You Begin”, you can start using HeatWave ML.
Proceed through the following steps to prepare data, train a model, make predictions, and generate explanations:
Prepare and load training and test data. See Section 3.3, “Preparing Data”.
Train a machine learning model. See Section 3.4, “Training a Model”.
Make predictions with test data using a trained model. See Section 3.6, “Predictions”.
Run explanations on test data using a trained model to understand how predictions are made. See Section 3.7, “Explanations”.
Score your machine learning model to assess its reliability. See Section 3.9.6, “Scoring Models”.
View a model explanation to understand how the model makes predictions. See Section 3.9.7, “Model Explanations”.
Alternatively, you can jump ahead to the Iris Data Set Machine Learning Quickstart, which provides a quick run-through of HeatWave ML capabilities using a simple, well-known machine learning data set. See Section 6.3, “Iris Data Set Machine Learning Quickstart”.