Class: TargetEncoder

Target Encoder for regression and classification targets.

Each category is encoded based on a shrunk estimate of the average target values for observations belonging to the category. The encoding scheme mixes the global target mean with the target mean conditioned on the value of the category (see [MIC]).

When the target type is “multiclass”, encodings are based on the conditional probability estimate for each class. The target is first binarized using the “one-vs-all” scheme via LabelBinarizer, then the average target value for each class and each category is used for encoding, resulting in n_features * n_classes encoded output features.

TargetEncoder considers missing values, such as np.nan or undefined, as another category and encodes them like any other category. Categories that are not seen during fit are encoded with the target mean, i.e. target_mean_.

For a demo on the importance of the TargetEncoder internal cross-fitting, see Target Encoder’s Internal Cross fitting. For a comparison of different encoders, refer to Comparing Target Encoder with Other Encoders. Read more in the User Guide.

Python Reference

Constructors

new TargetEncoder()

new TargetEncoder(opts?): TargetEncoder

Parameters

Parameter	Type	Description
`opts`?	`object`	-
`opts.categories`?	`"auto"`	Categories (unique values) per feature:
`opts.cv`?	`number`	Determines the number of folds in the cross fitting strategy used in `fit_transform`. For classification targets, `StratifiedKFold` is used and for continuous targets, `KFold` is used.
`opts.random_state`?	`number`	When `shuffle` is `true`, `random_state` affects the ordering of the indices, which controls the randomness of each fold. Otherwise, this parameter has no effect. Pass an int for reproducible output across multiple function calls. See Glossary.
`opts.shuffle`?	`boolean`	Whether to shuffle the data in `fit_transform` before splitting into folds. Note that the samples within each split will not be shuffled.
`opts.smooth`?	`number` \| `"auto"`	The amount of mixing of the target mean conditioned on the value of the category with the global target mean. A larger `smooth` value will put more weight on the global target mean. If `"auto"`, then `smooth` is set to an empirical Bayes estimate.
`opts.target_type`?	`"auto"` \| `"binary"` \| `"continuous"` \| `"multiclass"`	Type of target.

Returns TargetEncoder

Defined in generated/preprocessing/TargetEncoder.ts:29

Properties

Property	Type	Default value	Defined in
`_isDisposed`	`boolean`	`false`	generated/preprocessing/TargetEncoder.ts:27
`_isInitialized`	`boolean`	`false`	generated/preprocessing/TargetEncoder.ts:26
`_py`	`PythonBridge`	`undefined`	generated/preprocessing/TargetEncoder.ts:25
`id`	`string`	`undefined`	generated/preprocessing/TargetEncoder.ts:22
`opts`	`any`	`undefined`	generated/preprocessing/TargetEncoder.ts:23

Accessors

categories_

Get Signature

get categories_(): Promise<any>

The categories of each input feature determined during fitting or specified in categories (in order of the features in X and corresponding with the output of transform).

Returns Promise<any>

Defined in generated/preprocessing/TargetEncoder.ts:378

classes_

Get Signature

get classes_(): Promise<ArrayLike>

If target_type_ is ‘binary’ or ‘multiclass’, holds the label for each class, otherwise undefined.

Returns Promise<ArrayLike>

Defined in generated/preprocessing/TargetEncoder.ts:503

encodings_

Get Signature

get encodings_(): Promise<any[]>

Encodings learnt on all of X. For feature i, encodings_\[i\] are the encodings matching the categories listed in categories_\[i\]. When target_type_ is “multiclass”, the encoding for feature i and class j is stored in encodings_\[j + (i \* len(classes_))\]. E.g., for 2 features (f) and 3 classes (c), encodings are ordered: f0_c0, f0_c1, f0_c2, f1_c0, f1_c1, f1_c2,

Returns Promise<any[]>

Defined in generated/preprocessing/TargetEncoder.ts:353

feature_names_in_

Get Signature

get feature_names_in_(): Promise<ArrayLike>

Names of features seen during fit. Defined only when X has feature names that are all strings.

Returns Promise<ArrayLike>

Defined in generated/preprocessing/TargetEncoder.ts:478

n_features_in_

Get Signature

get n_features_in_(): Promise<number>

Number of features seen during fit.

Returns Promise<number>

Defined in generated/preprocessing/TargetEncoder.ts:453

py

Get Signature

get py(): PythonBridge

Returns PythonBridge

Set Signature

set py(pythonBridge): void

Parameters

Parameter	Type
`pythonBridge`	`PythonBridge`

Returns void

Defined in generated/preprocessing/TargetEncoder.ts:74

target_mean_

Get Signature

get target_mean_(): Promise<number>

The overall mean of the target. This value is only used in transform to encode categories.

Returns Promise<number>

Defined in generated/preprocessing/TargetEncoder.ts:428

target_type_

Get Signature

get target_type_(): Promise<string>

Type of target.

Returns Promise<string>

Defined in generated/preprocessing/TargetEncoder.ts:403

Methods

dispose()

dispose(): Promise<void>

Disposes of the underlying Python resources.

Once dispose() is called, the instance is no longer usable.

Returns Promise<void>

Defined in generated/preprocessing/TargetEncoder.ts:126

fit()

fit(opts): Promise<any>

Fit the TargetEncoder to X and y.

Parameters

Parameter	Type	Description
`opts`	`object`	-
`opts.X`?	`ArrayLike`[]	The data to determine the categories of each feature.
`opts.y`?	`ArrayLike`	The target data used to encode the categories.

Returns Promise<any>

Defined in generated/preprocessing/TargetEncoder.ts:143

fit_transform()

fit_transform(opts): Promise<ArrayLike[]>

Fit TargetEncoder and transform X with the target encoding.

Parameters

Parameter	Type	Description
`opts`	`object`	-
`opts.X`?	`ArrayLike`[]	The data to determine the categories of each feature.
`opts.y`?	`ArrayLike`	The target data used to encode the categories.