package lnsmote

import "git.sr.ht/~shulhan/pakakeh.go/lib/mining/resampling/lnsmote"

Package lnsmote implement the Local-Neighborhood algorithm from the paper,

Maciejewski, Tomasz, and Jerzy Stefanowski. "Local neighbourhood extension of SMOTE for mining imbalanced data." Computational Intelligence and Data Mining (CIDM), 2011 IEEE Symposium on. IEEE, 2011.

Index ¶

• type Runtime
• func New(percentOver, k, classIndex int, classMinor, outliers string) ( lnsmoteRun *Runtime, )
• func (in *Runtime) Init(dataset tabula.DatasetInterface)
• func (in *Runtime) Resampling(dataset tabula.DatasetInterface) ( e error, )

Types ¶

type Runtime ¶

type Runtime struct {

	// ClassMinor the minority sample in dataset that we want to
	// oversampling.
	ClassMinor string `json:"ClassMinor"`

	// OutliersFile if its not empty then outliers will be saved in file
	// specified by this option.
	OutliersFile string `json:"OutliersFile"`

	// Runtime of SMOTE, since this module extend the SMOTE method.
	smote.Runtime
	// contains filtered or unexported fields
}

Runtime parameters for input and output.

func New ¶

func New(percentOver, k, classIndex int, classMinor, outliers string) (
	lnsmoteRun *Runtime,
)

New create and return new LnSmote object.

func (*Runtime) Init ¶

func (in *Runtime) Init(dataset tabula.DatasetInterface)

Init will initialize LNSmote runtime by checking input values and set it to default if not set or invalid.

func (*Runtime) Resampling ¶

func (in *Runtime) Resampling(dataset tabula.DatasetInterface) (
	e error,
)

Resampling will run resampling process on dataset and return the synthetic samples.

Source Files ¶

lnsmote.go