GitHub
deepbox/preprocess

Encoders & Vectorizers

Categorical encoders and text vectorizers for tabular and text preprocessing pipelines.
Categorical
Text

LabelBinarizer

Binarize labels in a one-vs-all fashion.

LabelEncoder

Encode target labels with value between 0 and n_classes-1.

MultiLabelBinarizer

Transform multi-label classification data to binary format.

OneHotEncoder

Encode categorical features as one-hot numeric array.

OrdinalEncoder

Encode categorical features as integer array.

TargetEncoder

Target-based encoding for categorical features.

CountVectorizer

Convert a collection of text documents to a matrix of token counts.

HashingVectorizer

Convert text documents to a fixed-size feature matrix using the hashing trick.

TfidfVectorizer

Convert a collection of text documents to a TF-IDF feature matrix.

preprocess-encoders.ts
import {  CountVectorizer,  OneHotEncoder,  OrdinalEncoder,  TfidfVectorizer,} from "deepbox/preprocess";const oneHot = new OneHotEncoder();console.log(oneHot.fitTransform([["cat"], ["dog"], ["cat"]]).toString());const ordinal = new OrdinalEncoder();console.log(ordinal.fitTransform([["low"], ["medium"], ["high"]]).toString());const tfidf = new TfidfVectorizer();console.log(tfidf.fitTransformText(["deepbox docs", "deepbox ml"]).toString());console.log(new CountVectorizer().fitTransformText(["a b", "a a"]).toString());