SimpleImputer

Defined in: ds/src/ml/impute.js:148

Simple imputation strategies for missing values Compatible with sklearn.impute.SimpleImputer

Example

const imputer = new SimpleImputer({ strategy: 'mean' });
imputer.fit(X_train);
const X_filled = imputer.transform(X_test);

Constructors

Constructor

new SimpleImputer(options?): SimpleImputer

Defined in: ds/src/ml/impute.js:155

Parameters

options?

copy

boolean = true

If true, create copy of X (default: true)

fill_value

string | number = null

Value to use for ‘constant’ strategy

strategy

string = "mean"

‘mean’, ‘median’, ‘most_frequent’, or ‘constant’ (default: ‘mean’)

Returns

SimpleImputer

Properties

_groupModels

_groupModels: Map<any, any>

Defined in: ds/src/ml/impute.js:175

---

_originalData

_originalData: any

Defined in: ds/src/ml/impute.js:176

---

_tableColumns

_tableColumns: any

Defined in: ds/src/ml/impute.js:174

---

copy

copy: boolean

Defined in: ds/src/ml/impute.js:171

---

fill_value

fill_value: string | number

Defined in: ds/src/ml/impute.js:170

---

nFeatures_

nFeatures_: number

Defined in: ds/src/ml/impute.js:173

---

statistics_

statistics_: any[]

Defined in: ds/src/ml/impute.js:172

---

strategy

strategy: string

Defined in: ds/src/ml/impute.js:169

Methods

_fitSingleModel()

_fitSingleModel(X): any

Defined in: ds/src/ml/impute.js:262

Internal method to fit a single imputation model

Parameters

X

number[][]

2D array of numeric data

Returns

any

Model statistics

---

_transformWithModel()

_transformWithModel(X, statistics): number[][]

Defined in: ds/src/ml/impute.js:394

Internal method to transform data with specific statistics

Parameters

X

number[][]

Data to transform

statistics

number[]

Statistics to use for imputation

Returns

number[][]

Transformed data

---

fit()

fit(X): SimpleImputer

Defined in: ds/src/ml/impute.js:184

Fit the imputer on training data

Parameters

X

any

Training data, table object, or {data, columns, group} format

Returns

SimpleImputer

this

---

fit_transform()

fit_transform(X): number[][]

Defined in: ds/src/ml/impute.js:415

Fit and transform in one step

Parameters

X

any

Data to fit and transform

Returns

number[][]

Transformed data

---

transform()

transform(X): any[] | number[][]

Defined in: ds/src/ml/impute.js:311

Transform data by filling missing values

Parameters

X

any

Data to transform, table object, or {data, columns, group} format

Returns

any[] | number[][]

Transformed data (array if input was table)