Classification and regression overview

Binary classification models predict a binary outcome (one oftwo classes). Use this model type for yes or no questions. For example, you might wantto build a binary classification model to predict whether a customer wouldbuy a subscription. Generally, a binary classificationproblem requires less data than other model types.

Multi-class classification models predict one class from threeor more discrete classes. Use this model type for categorization. For example, as aretailer, you might want to build a multi-class classification model to segmentcustomers into different personas.

Regression models predict a continuous value. For example, as a retailer,you might want to build a regression model to predict how much acustomer will spend next month.

Workflow for creating a classification or regression model and making inferences

The process for creating a classification or regression model inVertex AI is as follows:

StepsDescription
1.Prepare training dataPrepare your training data for model training.
2.Create a datasetCreate a new dataset and associate your prepared training data with it.
3.Train a modelTrain a classification or regression model in Vertex AI using your dataset.
4.Evaluate your modelEvaluate your newly trained model for inference accuracy.
5.View model architectureView the hyperparameter logs of the tuning trials and the hyperparameter logs of the final model.
6. Get inferences from your model

To get real-time inferences, deploy your model andget online inferences.

If you don't need real-time inferences, makebatch inference requests directly to your model.

Except as otherwise noted, the content of this page is licensed under theCreative Commons Attribution 4.0 License, and code samples are licensed under theApache 2.0 License. For details, see theGoogle Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.