GAUSS Machine Learning (GML)#

Description#

Provides tools to model, analyze, and predict data using fundamental machine learning techniques.

Installation#

Please contact us directly to request a copy of the GAUSS Machine Learning library.

Requires GAUSS 22 or higher.

Commands#

binaryClassMetrics()

Computes statistics to assess the quality of binary predictions and prints out a report.

classificationMetrics()

Computes statistics to assess the quality of classification predictions and prints out a report.

cvSplit()

Returns the test and training set for the ith of k cross validation splits for a given set of dependent and independent variables.

decForestCFit()

Fit a decision forest classification model.

decForestPredict()

Predicts responses using the output from decForestCFit() or decForestRFit() and matrix of independent variables.

decForestRFit()

Fit a decision forest regression model.

kmeansFit()

Partitions data into k clusters, using the kmeans algorithm.

kmeansPredict()

Partitions data into k clusters, based upon k user supplied centroids.

knnFit()

Creates a K-D tree model from training data for efficient KNN predictions.

knnClassify()

Creates nearest neighbor predictions.

lassoFit()

Fit a linear model with an L1 penalty.

lmPredict()

Predict response using output from lassoFit(), ridgeFit(), or ridgeCFit() and matrix of independent variables.

logisticRegFit()

Fit a logistic regression model with an optional L1 and/or L2 penalty.

meanSquaredError()

Returns the mean squared error between two input vectors, or sets of vectors.

oneHot()

Returns a matrix of one-hot (indicator) variables from a vector or dataframe column.

pcaFit()

Performs principal component dimension reduction.

pcaTransform()

Reduces the dimension of a matrix using principal component vectors previously returned by pcaFit().

pcaTransformInv()

Transforms a matrix back to the original feature space of the X which was input to pcaFit().

plotClasses()

Plots predicted classes in color coded scatter plot.

plotLR()

Plots parameter path and mse path over regularization path after lassoFit() or ridgeFit().

plotVariableImportance()

Plots variable importance table after decision forest regression.

ridgeFit()

Fit a linear model with an L2 penalty.

ridgeCFit()

Fit a binary classification model with an L2 penalty.

ridgeCPredict()

Predict binary classifications from ridgeCFit() model.

splitData()

Returns test and training splits for a single matrix of variables.

trainTestSplit()

Returns test and training splits for a given set of dependent and independent variables.