@neurova/ai-data

Built-in datasets and CSV utilities. JS/TS port of neurova.datasets / sklearn.datasets-style loaders.

import { loadIris, asMatrix, parseCsv } from '@neurova/ai-data'
import { KNearestNeighbors, accuracyScore, trainTestSplit } from '@neurova/ai-ml'

const iris = loadIris()
const X = asMatrix(iris)
const y = Array.from(iris.target)
const { XTrain, XTest, yTrain, yTest } = trainTestSplit(X, y, 0.2, 42)
const knn = new KNearestNeighbors(3)
knn.train(XTrain, yTrain)
console.log(accuracyScore(yTest, knn.infer(XTest)))

Includes:

• loadIris() — inlined 150 × 4 dataset, no network required.
• loadFashionMnist({ subset: 'train' | 'test' }) — fetches gzipped IDX files via fetch + DecompressionStream (Node ≥ 18 / modern browsers).
• parseCsv(text, { header, delimiter }) — RFC-4180-ish CSV parser.
• Helpers: row, asMatrix.

© @analyticswithharry and Squid Consultancy Group Ltd. MIT.