Linear.LogisticRegression

interface LogisticRegressionProps {
    learningRate?: number;
    maxIter?: number;
}
constructor(props: LogisticRegressionProps = {})

Performs binary logistic regression optimized with gradient descent. The learningRate controls the step size during optimization and maxIter specifies the number of iterations.

Methods

• fit(trainX: number[][], trainY: number[]): void
• predict(testX: number[][]): number[]

fit

• trainX - Training features with shape [nSamples, nFeatures].
• trainY - Binary labels for each training sample.

predict

• testX - Feature matrix to classify.

Example

const clf = new LogisticRegression();
clf.fit(trainX, trainY);
const result = clf.predict(testX);

Practical guide: LogisticRegression in JavaScript and TypeScript

LogisticRegression estimates class probabilities for binary classification with stable, interpretable coefficients.

When to use LogisticRegression

• You need calibrated probability scores for threshold-based decisions.
• Binary outcomes are influenced by approximately linear decision boundaries.
• You want a strong and explainable baseline in TypeScript projects.

Implementation workflow

1. Standardize inputs and confirm label quality for binary classes.
2. Fit with appropriate iteration settings and monitor convergence.
3. Tune decision threshold based on business costs of false positives/negatives.

JavaScript deployment notes

• Prefer feature scaling for distance-based and gradient-based algorithms to improve stability.
• In browser apps, run heavy training in Web Workers to keep UI interactions smooth.
• Keep a simple baseline from the same module as a fallback model for comparison.

Search intents this page targets

• LogisticRegression JavaScript
• LogisticRegression TypeScript
• LogisticRegression browser machine learning
• @kanaries/ml LogisticRegression