Regression overview
A common use case for machine learning is predicting the value of a numerical metric for new data by using a model trained on similar historical data. For example, you might want to predict a house's expected sale price. By using the house's location and characteristics as features, you can compare this house to similar houses that have already sold, and use their sales prices to estimate the house's sale price.
You can use any of the following models in combination with the ML.PREDICT function to perform regression:
- Linear regression models: use linear regression by setting the
MODEL_TYPEoption toLINEAR_REG. - Boosted tree models: use a gradient boosted decision tree by setting the
MODEL_TYPEoption toBOOSTED_TREE_REGRESSOR. - Random forest models: use a random forest by setting the
MODEL_TYPEoption toRANDOM_FOREST_REGRESSOR. - Deep neural network (DNN) models: use a neural network by setting the
MODEL_TYPEoption toDNN_REGRESSOR. - Wide & Deep models: use wide & deep learning by setting the
MODEL_TYPEoption toDNN_LINEAR_COMBINED_REGRESSOR. - AutoML models: use an AutoML classification model by setting the
MODEL_TYPEoption toAUTOML_REGRESSOR.
Recommended knowledge
By using the default settings in the CREATE MODEL statements and the ML.PREDICT function, you can create and use a regression model even without much ML knowledge. However, having basic knowledge about ML development helps you optimize both your data and your model to deliver better results. We recommend using the following resources to develop familiarity with ML techniques and processes: