Linear Regression

import { tensor } from "deepbox/ndarray";
import { trainTestSplit } from "deepbox/preprocess";

// Generate data: y = 2x + 3 + noise
const X_data: number[][] = [];
const y_data: number[] = [];

for (let i = 0; i < 100; i++) {
  const x = i / 10;
  const y = 2 * x + 3 + (Math.random() - 0.5) * 2;
  X_data.push([x]);
  y_data.push(y);
}

const X = tensor(X_data);
const y = tensor(y_data);

console.log(`Dataset: ${X.shape[0]} samples, ${X.shape[1]} features`);

Dataset: 100 samples, 1 features

const [X_train, X_test, y_train, y_test] = trainTestSplit(X, y, {
  testSize: 0.2,
  randomState: 42,
});

console.log(`Training set: ${X_train.shape[0]} samples`);
console.log(`Test set: ${X_test.shape[0]} samples`);

Training set: 80 samples
Test set: 20 samples

import { LinearRegression } from "deepbox/ml";

const model = new LinearRegression();
model.fit(X_train, y_train);

console.log("Model trained!");
console.log(`Coefficients: ${model.coef?.toString()}`);
console.log(`Intercept: ${model.intercept}`);

Model trained!
Coefficients: Tensor([2.0134])
Intercept: 2.9876

import { mae, mse, r2Score } from "deepbox/metrics";

const y_pred = model.predict(X_test);

const r2 = r2Score(y_test, y_pred);
const mseValue = mse(y_test, y_pred);
const maeValue = mae(y_test, y_pred);

console.log("Model Performance:");
console.log(`R² Score: ${r2.toFixed(4)}`);
console.log(`MSE: ${mseValue.toFixed(4)}`);
console.log(`MAE: ${maeValue.toFixed(4)}`);

Model Performance:
R² Score: 0.9912
MSE: 0.3421
MAE: 0.4856