Class Lasso

Least absolute shrinkage and selection operator

Index

Constructors

Properties

_lambda _method _w

Methods

_calc_b0 _cd _ista _lars _soft_thresholding fit importance predict

Constructors

constructor

new Lasso(lambda?: number, method?: "ISTA" | "CD" | "LARS"): Lasso
Parameters
- Optionallambda: number
  Regularization strength
- Optionalmethod: "ISTA" | "CD" | "LARS"
  Method name
Returns Lasso
- Defined in lib/model/lasso.d.ts:9

Properties

_lambda

_lambda: number

_method

_method: "ISTA" | "CD" | "LARS"

_w

_w: any

Methods

_calc_b0

_calc_b0(x: any, y: any): void
Parameters
- x: any
- y: any
Returns void
- Defined in lib/model/lasso.d.ts:14

_cd

_cd(x: any, y: any): void
Parameters
- x: any
- y: any
Returns void
- Defined in lib/model/lasso.d.ts:22

_ista

_ista(x: any, y: any): void
Parameters
- x: any
- y: any
Returns void
- Defined in lib/model/lasso.d.ts:21

_lars

_lars(x: any, y: any): void
Parameters
- x: any
- y: any
Returns void
- Defined in lib/model/lasso.d.ts:23

_soft_thresholding

_soft_thresholding(x: any, l: any): void
Parameters
- x: any
- l: any
Returns void
- Defined in lib/model/lasso.d.ts:13

fit

fit(x: number[][], y: number[][]): void
Fit model.
Parameters
- x: number[][]
  Training data
- y: number[][]
  Target values
Returns void
- Defined in lib/model/lasso.d.ts:20

importance

importance(): number[]
Returns importances of the features.

Returns number[]
Importances
- Defined in lib/model/lasso.d.ts:34

predict

predict(x: number[][]): number[][]
Returns predicted values.
Parameters
- x: number[][]
  Sample data
Returns number[][]
Predicted values
- Defined in lib/model/lasso.d.ts:29