Accelerating the pace of engineering and science

Statistics and Machine Learning Toolbox

Machine Learning for High-Dimensional Data

Machine Learning for High-Dimensional Data

Perform fast fitting of linear classification and regression models with techniques such as stochastic gradient descent and (L)BFGS using fitclinear and fitrlinear functions

Classification Learner

Classification Learner

Train multiple models automatically, visualize results by class labels, and perform logistic regression classification

Performance

Performance

Perform clustering using kmeans, kmedoids, and Gaussian mixture models faster when data has a large number of clusters

Probability Distributions

Probability Distributions

Fit kernel smoothing density to multivariate data using the ksdensity and mvksdensity functions

Stable Distributions

Stable Distributions

Model financial and other data that requires heavy-tailed distributions

Latest Releases

R2016a (Version 10.2) - 3 Mar 2016

Version 10.2, part of Release 2016a, includes the following enhancements:

  • Machine Learning for High-Dimensional Data: Perform fast fitting of linear classification and regression models with techniques such as stochastic gradient descent and (L)BFGS using fitclinear and fitrlinear functions
  • Classification Learner: Train multiple models automatically, visualize results by class labels, and perform logistic regression classification
  • Performance: Perform clustering using kmeans, kmedoids, and Gaussian mixture models faster when data has a large number of clusters
  • Probability Distributions: Fit kernel smoothing density to multivariate data using the ksdensity and mvksdensity functions
  • Stable Distributions: Model financial and other data that requires heavy-tailed distributions

See the Release Notes for details.

R2015b (Version 10.1) - 3 Sep 2015

Version 10.1, part of Release 2015b, includes the following enhancements:

  • Classification Learner: Train discriminant analysis to classify data, train models using categorical predictors, and perform dimensionality reduction using PCA
  • Nonparametric Regression: Fit models using support vector regression (SVR) or Gaussian processes (Kriging)​
  • Tables and Categorical Data for Machine Learning: Use table and categorical predictors in classification and nonparametric regression functions and in Classification Learner​
  • Code Generation: Automatically generate C and C++ code for kmeans and randsample functions (using MATLAB Coder)​
  • GPU Acceleration: Speed up computation for over 65 functions including probability distributions, descriptive statistics, and hypothesis testing (using Parallel Computing Toolbox)

See the Release Notes for details.

R2015a (Version 10.0) - 5 Mar 2015

Version 10.0, part of Release 2015a, includes the following enhancements:

  • Classification app to train models and classify data using supervised machine learning
  • Statistical tests for comparing accuracies of two classification models using compareHoldout, testcholdout, and testckfold functions
  • Speedup of kmedoids, fitcknn, and other functions when using cosine, correlation, or spearman distance calculations
  • Performance enhancements for decision trees and performance curves​​
  • Additional option to control decision tree depth using 'MaxNumSplits' argument in fitctree, fitrtree, and templateTree functions
  • Code generation for pca and probability distribution functions (using MATLAB Coder)
  • Power and sample size for two-sample t-test using sampsizepwr function

See the Release Notes for details.

R2014b (Version 9.1) - 2 Oct 2014

Version 9.1, part of Release 2014b, includes the following enhancements:

  • Multiclass learning for support vector machines and other classifiers using the fitcecoc function
  • Generalized linear mixed-effects models using the fitglme function
  • Clustering that is robust to outliers using the kmedoids function
  • Speedup of the kmeans and gmdistribution clustering using the kmeans++ algorithm
  • Fisher's exact test for 2-by-2 contingency tables

See the Release Notes for details.

R2014a (Version 9.0) - 6 Mar 2014

Version 9.0, part of Release 2014a, includes the following enhancements:

  • Repeated measures modeling for data with multiple measurements per subject
  • fitcsvm function for enhanced performance of support vector machines (SVMs) for binary classification
  • evalclusters methods to expand the number of clusters and number of gap criterion simulations
  • p-value output from the multcompare function
  • mnrfit, lassoglm, and fitglm functions accept categorical variables as responses
  • Functions accept table inputs as an alternative to dataset array inputs
  • Functions and model properties return a table rather than a dataset array

See the Release Notes for details.