API Reference - Models - BayesianLinearRegression
BayesianLinearRegression is a supervised machine learning model that predicts continuous values (e.g. 1.2, -32, 90, -1.2 and etc. ). It uses matrix calculations to find the best model parameters.
Stored Model Parameters
Contains a matrix.
-
ModelParameters[1][I][J]: posteriorMeanVector. Value of matrix at row I and column J. The rows are the features.
-
ModelParameters[2][I][J]: posteriorCovarianceMatrix. Value of matrix at row I and column J. The rows and columns are the features.
Constructors
new()
Create new model object. If any of the arguments are nil, default argument values for that argument will be used.
BayesianLinearRegression.new(priorPrecision: number, likelihoodPrecision: number, useLogProbabilities: boolean): ModelObject
Parameters:
-
priorPrecision: How confident we are with the previous model weights. Higher values means more confidence in the previous model weights. [Default: 1]
-
likelihoodPrecision: How reliable are the observations. Higher values means observations are more reliable. [Default: 1]
-
useLogProbabilities: Set whether or not to use log probabilities when generating predicted probabilities in predict() function. [Default: False]
Returns:
- ModelObject: The generated model object.
Functions
train()
Train the model.
BayesianLinearRegression:train(featureMatrix: Matrix, labelVector: Matrix)
Parameters:
-
featureMatrix: Matrix containing all data.
-
labelVector: A (n x 1) matrix containing values related to featureMatrix.
predict()
Predict the value for a given data.
BayesianLinearRegression:predict(featureMatrix: Matrix, thresholdMatrix: Matrix): Matrix -OR- Matrix, Matrix
Parameters:
-
featureMatrix: Matrix containing data.
-
thresholdMatrix: A matrix of threshold values for computing predicted probabilities. If provided, the model returns both predicted values and the probability that the prediction exceeds the threshold(s).
Returns:
- predictedVector: A vector containing values that are predicted by the model.
– OR –
-
predictedVector: A vector containing values that are predicted by the model.
-
predictedProbabilityMatrix: A matrix contining the probability of the values with the given threshold.