CrossEntropyLoss is the loss for LogisticRegression and Classifiers J: -y*math.Log(h)-(1.-y)*log(1.-h) grad: hprime*(-y/h + (1-y)/(1-h)) .

EnetPath Compute elastic net path with coordinate descent no preprocessing is done here, you must have called PreprocessData before.

LassoPath Compute lasso path with coordinate descent.

LinFit is an internal helper to fit linear regressions.

LinFitGOM fits a regression with a gonum/optimizer Method.

LogLoss for one versus rest classifiers.

NewBayesianRidge creates a *BayesianRidge with defaults.

NewElasticNet creates a *ElasticNet with Alpha=1 and L1Ratio=0.5.

NewLasso creates a *ElasticNetRegression with Alpha=1 and L1Ratio = 1.

NewLinearRegression create a *LinearRegression with defaults implemented as mat.Dense.Solve.

NewLogisticRegression returns a LogisticRegression with defaults: Alpha=1/C=1; Tol=1e-4.

NewMultiTaskElasticNet creates a *ElasticNet with Alpha=1 and L1Ratio=0.5.

NewMultiTaskLasso creates a *RegularizedRegression with Alpha=1 and L1Ratio=1.

NewRidge creates a *RegularizedRegression with Alpha=1.

NewSGDRegressor creates a *SGDRegressor with defaults.

PreprocessData center and normalize data.

SquareLoss Quadratic Loss, for regressions Ytrue, X, Theta must be passed in Ypred,Ydiff,Ytmp are temporary matrices passed in here to avoid reallocations.

LossFunctions is the map of implemented loss functions.

BayesianRidge regression struct.

CDResult is the coordinate descent specific part in the regression result.

ElasticNet is the struct for coordinate descent regularized regressions: ElasticNet,Ridge,Lasso Selection is cyclic or random.

LinearModel is a base struct for multioutput regressions.

LinearRegression ia Ordinary least squares Linear Regression.

LinFitOptions are options for LinFit.

LinFitResult is the result or LinFit.

LogisticRegression Logistic Regression (aka logit, MaxEnt) classifier.

RegularizedRegression is a common structure for ElasticNet,Lasso and Ridge.

SGDRegressor base struct should be named GonumOptimizeRegressor implemented as a per-output optimization of (possibly regularized) square-loss with gonum/optimize methods.

Activation is borrowed from base package.

Lasso is an alias for ElasticNet.

Loss puts cost in J and cost gradient in grad.

MultiTaskElasticNet is an alias for ElasticNet.

MultiTaskLasso is an alias for ElasticNet/Lasso.

Ridge is an alias for RegularizedRegression.