Train Regression Models in Regression Learner App

You can use Regression Learner to train regression models including linear regression models, regression trees, Gaussian process regression models, support vector machines, ensembles of regression trees, and neural network regression models. In addition to training models, you can explore your data, select features, specify validation schemes, and evaluate results. You can export a model to the workspace to use the model with new data or generate MATLAB^®code to learn about programmatic regression.

Training a model in Regression Learner consists of two parts:

Validated Model: Train a model with a validation scheme. By default, the app protects against overfitting by applying cross-validation. Alternatively, you can choose holdout validation. The validated model is visible in the app.
Full Model: Train a model on full data without validation. The app trains this model simultaneously with the validated model. However, the model trained on full data is not visible in the app. When you choose a regression model to export to the workspace, Regression Learner exports the full model.

The app displays the results of the validated model. Diagnostic measures, such as model accuracy, and plots, such as a response plot or residuals plot, reflect the validated model results. You can automatically train one or more regression models, compare validation results, and choose the best model that works for your regression problem. When you choose a model to export to the workspace, Regression Learner exports the full model. Because Regression Learner creates a model object of the full model during training, you experience no lag time when you export the model. You can use the exported model to make predictions on new data.

To get started by training a selection of model types, see自动回归模型训练. If you already know which regression model you want to train, seeManual Regression Model Training.

自动回归模型训练

You can use Regression Learner to automatically train a selection of different regression models on your data.

Get started by automatically training multiple models simultaneously. You can quickly try a selection of models, and then explore promising models interactively.
If you already know what model type you want, then you can train individual models instead. SeeManual Regression Model Training.

On theAppstab, in theMachine Learning and Deep Learninggroup, click回归的学习者.
ClickNew Sessionand select data from the workspace or from a file. Specify a response variable and variables to use as predictors. SeeSelect Data and Validation for Regression Problem.
On the回归的学习者tab, in theModel Typesection, click the arrow to expand the list of regression models. SelectAll Quick-To-Train. This option trains all the model presets that are fast to fit.
ClickTrain.

Note

If you have Parallel Computing Toolbox™, you can train the models in parallel. SeeParallel Regression Model Training.

A selection of model types appears in theModelspane. When the models finish training, the bestRMSE (Validation)score is highlighted in a box.
Click models in theModelspane and open the corresponding plots to explore the results.
For the next steps, seeManual Regression Model TrainingorCompare and Improve Regression Models.
To try all the nonoptimizable model presets available, clickAll, and then clickTrain.

Manual Regression Model Training

To explore individual model types, you can train models one at a time or train a group of models of the same type.

Choose a model type. On the回归的学习者tab, in theModel Typesection, click a model type. To see all available model options, click the arrow in theModel Typesection to expand the list of regression models. The nonoptimizable model options in the gallery are preset starting points with different settings, suitable for a range of different regression problems.
To read descriptions of the models, switch to the details view or place the mouse over a button to display its tooltip.

For more information on each option, seeChoose Regression Model Options.
After selecting a model, clickTrain.
Repeat to explore different models.

Tip

Select regression trees first. If your trained models do not predict the response accurately enough, then try other models with higher flexibility. To avoid overfitting, look for a less flexible model that provides sufficient accuracy.
If you want to try all nonoptimizable models of the same or different types, then select one of theAlloptions in the gallery.
Alternatively, if you want to automatically tune hyperparameters of a specific model type, select the corresponding的优化zablemodel and perform hyperparameter optimization. For more information, seeHyperparameter Optimization in Regression Learner App.

For next steps, seeCompare and Improve Regression Models.

Parallel Regression Model Training

You can train models in parallel using Regression Learner if you have Parallel Computing Toolbox. Parallel training allows you to train multiple models simultaneously and continue working.

To control parallel training, toggle theUse Parallelbutton on the app toolstrip. TheUse Parallelbutton is available only if you have Parallel Computing Toolbox.

The first time you clickTrainafter clicking theUse Parallelbutton, a dialog box is displayed while the app opens a parallel pool of workers. After the pool opens, you can train multiple models at once.

When models are training in parallel, progress indicators appear on each training and queued model in theModelspane. If you want, you can cancel individual models. During training, you can examine results and plots from models, and initiate training of more models.

If you have Parallel Computing Toolbox, then parallel training is available in Regression Learner, and you do not need to set theUseParalleloption of thestatsetfunction.

Note

You cannot perform hyperparameter optimization in parallel. The app disables theUse Parallelbutton when you select an optimizable model. If you then select a nonoptimizable model, the button is off by default.

Compare and Improve Regression Models

Examine theRMSE (Validation)score reported in theModelspane for each model. Click models in theModelspane and open the corresponding plots to explore the results. Compare model performance by inspecting results in the plots. You can rearrange the layout of the plots to compare results across multiple models: use the options in theLayoutbutton, drag and drop plots, or select the options provided by the Document Actions arrowlocated to the right of the model plot tabs.
Additionally, you can compare the models by using theSort byoptions in theModelspane. Delete any unwanted model by selecting the model and clicking theDelete selected modelbutton in the upper right of the pane, or right-clicking the model and selectingDelete model.
SeeAssess Model Performance in Regression Learner.
Select the best model in theModelspane and then try including and excluding different features in the model. ClickFeature Selection.
Try the response plot to help you identify features to remove. See if you can improve the model by removing features with low predictive power. Specify predictors to include in the model, and train new models using the new options. Compare results among the models in theModelspane.
You also can try transforming features with PCA to reduce dimensionality.
SeeFeature Selection and Feature Transformation Using Regression Learner App.
Improve the model further by changing model parameter settings in the Advanced dialog box. Then, train using the new options. To learn how to control model flexibility, seeChoose Regression Model Options. For information on how to tune model parameter settings automatically, seeHyperparameter Optimization in Regression Learner App.
If feature selection, PCA, or new parameter settings improve your model, try trainingAll模型类型与新设置。看看另一个model type does better with the new settings.

Tip

To avoid overfitting, look for a less flexible model that provides sufficient accuracy. For example, look for simple models, such as regression trees that are fast and easy to interpret. If your models are not accurate enough, then try other models with higher flexibility, such as ensembles. To learn about the model flexibility, seeChoose Regression Model Options.

This figure shows the app with aModelspane containing various regression model types.

For a step-by-step example comparing different regression models, seeTrain Regression Trees Using Regression Learner App.

Next, you can generate code to train the model with different data or export trained models to the workspace to make predictions using new data. SeeExport Regression Model to Predict New Data.