Safe Haskell	None
Language	Haskell2010

DataFrame.LinearModel

Description

Linear regression (OLS, ridge, lasso, elastic net) and logistic models.

Synopsis

Documentation

Hyper-parameters for the FISTA solver.

Constructors

Fields

scL1Lambda :: !Double
Strength of the L1 penalty on weights (intercept is not regularized).
scL2Lambda :: !Double
Strength of the L2 penalty (elastic net); 0 reduces to pure L1.
scMaxIter :: !Int
Maximum number of FISTA iterations.
scTol :: !Double
Convergence tolerance on the weight delta (L-inf norm).
scSampleWeights :: !(Maybe (Vector Double))
Optional per-row sample weights with mean 1 (Nothing is uniform).

Instances details

Show SolverConfig
Instance details Defined in DataFrame.LinearSolver Methods showsPrec :: Int -> SolverConfig -> ShowS # show :: SolverConfig -> String # showList :: [SolverConfig] -> ShowS #
Eq SolverConfig
Instance details Defined in DataFrame.LinearSolver Methods (==) :: SolverConfig -> SolverConfig -> Bool # (/=) :: SolverConfig -> SolverConfig -> Bool #

A fitted linear classifier; zero weights mark features dropped by L1.

Constructors

LinearModel
Fields lmWeights :: !(Vector Double) lmIntercept :: !Double lmFeatureNames :: !(Vector Text)

Instances details

Show LinearModel
Instance details Defined in DataFrame.LinearSolver Methods showsPrec :: Int -> LinearModel -> ShowS # show :: LinearModel -> String # showList :: [LinearModel] -> ShowS #
Eq LinearModel
Instance details Defined in DataFrame.LinearSolver Methods (==) :: LinearModel -> LinearModel -> Bool # (/=) :: LinearModel -> LinearModel -> Bool #