Statistics and Machine Learning Toolbox Apps
With Statistics and Machine Learning Toolbox apps, you can perform common statistics and machine learning tasks through an interactive interface, without writing any code. The Classification Learner, Regression Learner, and Distribution Fitting apps are included in the toolbox and let you fit distributions and train models for classification and regression. You can find them, along with the apps for your other installed products, in the Apps tab on the MATLAB Toolstrip. In addition, you can find many apps on the MATLAB File Exchange that have been authored by the MATLAB User Community using the App Designer.
The Classification Learner app lets you train models to classify data.
Using Classification Learner, you can perform common machine learning tasks such as interactively exploring your data, selecting features, specifying validation schemes, training models, and assessing results. Choose from many popular classification models including decision trees, logistic regression, support vector machines (SVM), and ensemble classifiers that use bagging, boosting, or random subspace methods.
Classification Learner helps you choose the best model for your data by letting you perform model assessment and model comparisons using confusion matrices and ROC curves. Export classification models to the MATLAB workspace to generate predictions on new data, generate MATLAB code to train the same model on new data, or further optimize it by applying automated feature selection or hyperparameter tuning.
The Regression Learner app lets you train models to predict continuous-valued responses using regression methods.
Just like Classification Learner, Regression Learner lets you perform all common machine learning tasks for regression models and export your model or generate MATLAB code for further evaluations and optimization. Choose from regression models including linear regression models, support vector machines (SVM), Gaussian process regression models, and ensembles of trees.
To help you choose the best model, use common regression visualizations such as plots of the response or the residuals, or compare predicted versus actual responses.
You can use complete data or censored (reliability) data, and you can exclude data, save and load sessions, and generate MATLAB code.