Objeck Documentation

All Bundles

Bundle Core machine learning types: the seedable Random generator, Matrix2D linear algebra and activations, matrix reference holders, the NeuralNetwork and CSV-backed MatrixReader. Compile with -lib ml.

ElasticNet

Elastic-net regularized linear regression fitted by cyclic coordinate descent on mean-centered data with an unpenalized intercept. Minimizes (1/2n)||y - Xw - b||^2 + alpha*(l1_ratio*|w|_1 + (1-l1_ratio)/2*|w|_2^2). l1_ratio = 1.0 is the lasso; l1_ratio = 0.0 is ridge (use RidgeRegression for the closed-form solution). The L1 term drives uninformative feature coefficients exactly to zero.

Example

model := ElasticNet->New(0.1, 0.5);
model->Fit(X, y);
preds := model->Predict(X);

Fit #

Fits the model using cyclic coordinate descent.

method : public : Fit(X:Float[,], y:Float[,]) ~ Bool

Parameters

Name	Type	Description
X	Float	feature matrix (rows=samples, cols=features)
y	Float	target matrix (rows=samples, cols=1)

Return

Type	Description
Bool	true if fitting succeeded

GetCoefficients #

Gets the fitted feature coefficients (the intercept is separate).

method : public : GetCoefficients() ~ Float[]

Return

Type	Description
Float	coefficient array, or Nil if not fitted

GetIntercept #

Gets the fitted intercept term.

method : public : GetIntercept() ~ Float

Return

Type	Description
Float	intercept

GetMSE #

Gets the mean squared error over the training data.

method : public : GetMSE() ~ Float

Return

Type	Description
Float	MSE

GetRMSE #

Gets the root mean squared error over the training data.

method : public : GetRMSE() ~ Float

Return

Type	Description
Float	RMSE

GetRSquared #

Gets the R-squared over the training data.

method : public : GetRSquared() ~ Float

Return

Type	Description
Float	R-squared value

IsFitted #

Whether the model has been fitted.

method : public : IsFitted() ~ Bool

Return

Type	Description
Bool	true if fitted

Load # function

Loads a fitted model from a file.

function : Load(filename:String) ~ ElasticNet

Parameters

Name	Type	Description
filename	String	file to load from

Return

Type	Description
ElasticNet	fitted model, or Nil on failure

New # constructor

Constructor with default optimization settings (1000 iterations, tolerance 0.0001).

New(alpha:Float, l1_ratio:Float)

Parameters

Name	Type	Description
alpha	Float	overall regularization strength (>= 0.0)
l1_ratio	Float	mix between L1 (1.0) and L2 (0.0) penalties

New # constructor

Constructor with explicit optimization settings.

New(alpha:Float, l1_ratio:Float, iterations:Int, tolerance:Float)

Parameters

Name	Type	Description
alpha	Float	overall regularization strength (>= 0.0)
l1_ratio	Float	mix between L1 (1.0) and L2 (0.0) penalties
iterations	Int	maximum coordinate-descent sweeps
tolerance	Float	stop when no coefficient moves more than this in a sweep

Predict #

Predicts target values for new input data.

method : public : Predict(X:Float[,]) ~ Float[,]

Parameters

Name	Type	Description
X	Float	feature matrix

Return

Type	Description
Float	prediction matrix, or Nil if not fitted

Score #

Scores the model on the given data (R-squared of predictions).

method : public : Score(X:Float[,], y:Float[,]) ~ Float

Parameters

Name	Type	Description
X	Float	feature matrix
y	Float	true target matrix (rows x 1)

Return

Type	Description
Float	R-squared, or 0.0 if not fitted or inputs are invalid

Store #

Saves the fitted model to a file.

method : public : Store(filename:String) ~ Bool

Parameters

Name	Type	Description
filename	String	file to store to

Return

Type	Description
Bool	true if successful, false otherwise