@neurova/ai-ml
v2.2.2
Published
neurova AI ML — classical learners with .train()/.infer() (port of Python nalyst).
Maintainers
analyticswithharryReadme
@neurova/ai-ml
Classical ML learners with the .train() / .infer() API.
JS/TS port of the nalyst Python library.
import { LogisticLearner, trainTestSplit, accuracyScore } from '@neurova/ai-ml'
const { Xtrain, Xtest, ytrain, ytest } = trainTestSplit(X, y, { testRatio: 0.2 })
const model = new LogisticLearner({ strength: 0.5 }).train(Xtrain, ytrain)
const acc = accuracyScore(ytest, model.infer(Xtest))Currently shipped: KNearestNeighbors, LogisticLearner, LinearRegression,
KMeans, trainTestSplit, accuracyScore, meanSquaredError,
BaseLearner, duplicate.
© @analyticswithharry and Squid Consultancy Group Ltd. MIT.