From e4ab0fb0f79ac973982e9afe2687c4fc7e03ed17 Mon Sep 17 00:00:00 2001 From: Vamsi-2581 <245859299+Vamsi-2581@users.noreply.github.com> Date: Tue, 20 Oct 2026 08:07:18 +0530 Subject: [PATCH] test: cover linear regression algorithm (#12) Signed-off-by: Vamsi-1 <245859299+Vamsi-2581@users.noreply.github.com> Assisted-by: OpenAI Codex --- docs/DEVELOPER_GUIDE.md | 4 +- tests/linear-regression-algorithm.test.ts | 60 +++++++++++++++++++++++ 2 files changed, 62 insertions(+), 2 deletions(-) create mode 100644 tests/linear-regression-algorithm.test.ts diff --git a/docs/DEVELOPER_GUIDE.md b/docs/DEVELOPER_GUIDE.md index 5bd4bda..2b62fb5 100644 --- a/docs/DEVELOPER_GUIDE.md +++ b/docs/DEVELOPER_GUIDE.md @@ -705,10 +705,10 @@ const handleInput = (event: Event) => { ### Unit Testing -Use Vitest for unit tests: +Unit tests live in the root `tests/` directory, with one test file per algorithm. Use Vitest for unit tests: ```typescript -// algorithms/__tests__/ExampleAlgorithm.test.ts +// tests/example-algorithm.test.ts import { describe, it, expect } from 'vitest'; import { ExampleAlgorithm } from '../ExampleAlgorithm'; diff --git a/tests/linear-regression-algorithm.test.ts b/tests/linear-regression-algorithm.test.ts new file mode 100644 index 0000000..152659d --- /dev/null +++ b/tests/linear-regression-algorithm.test.ts @@ -0,0 +1,60 @@ +/** + * Linear Regression Algorithm Tests + * + * Tests deterministic training, predictions, metrics, and visualization output. + */ + +import { describe, it, expect } from 'vitest'; +import { LinearRegression } from '../algorithms/regression/LinearRegression'; + +const X = [[1], [2], [3], [4]]; +const y = [2, 4, 6, 8]; + +function createAlgorithm(): LinearRegression { + return new LinearRegression({ learningRate: 0.1, epochs: 500, normalize: false }); +} + +describe('Linear Regression Algorithm', () => { + it('fits a known linear relationship and predicts unseen values', async () => { + const algorithm = createAlgorithm(); + + await algorithm.fit(X, y); + + expect(algorithm.predict([[5]])[0]).toBeCloseTo(10, 1); + algorithm.predict(X).forEach((prediction, index) => { + expect(prediction).toBeCloseTo(y[index], 1); + }); + }); + + it('reports regression metrics for the fitted training data', async () => { + const algorithm = createAlgorithm(); + await algorithm.fit(X, y); + + const metrics = algorithm.getMetrics(); + + expect(metrics.mse).toBeDefined(); + expect(metrics.rmse).toBeDefined(); + expect(metrics.mae).toBeDefined(); + expect(metrics.r2Score).toBeDefined(); + expect(metrics.mse!).toBeLessThan(0.01); + expect(metrics.mae!).toBeLessThan(0.1); + expect(metrics.r2Score!).toBeGreaterThan(0.99); + }); + + it('provides regression, residual, and loss visualizations after training', async () => { + const algorithm = createAlgorithm(); + await algorithm.fit(X, y); + + const visualizations = algorithm.getVisualizationData(); + + expect(visualizations).toHaveLength(3); + expect(visualizations[0].type).toBe('scatter2d'); + expect(visualizations[0].data.scatter.x).toEqual([1, 2, 3, 4]); + expect(visualizations[0].data.scatter.y).toEqual(y); + expect(visualizations[0].data.line.x).toHaveLength(101); + expect(visualizations[1].data.residuals.x).toEqual([1, 2, 3, 4]); + expect(visualizations[2].type).toBe('line'); + expect(visualizations[2].data.x).toHaveLength(500); + expect(visualizations[2].data.y).toHaveLength(500); + }); +});