npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2024 – Pkg Stats / Ryan Hefner

@andypai/neuroflow

v0.0.3

Published

simple neural network library inspired by karpathy/micrograd and tfjs

Downloads

10

Readme

NeuroFlow

NPM Version License

NeuroFlow is a JavaScript library that allows you to implement neural network layers and architectures from scratch, similar to the Linear layer in PyTorch. It provides a foundation for reimplementing arbitrary parts of other libraries like PyTorch and TensorFlow, and includes a host module to execute the code as part of a neural network.

This project started as a port of karpathy/micrograd to JavaScript for educational purposes. The API has been tweaked to prioritize readability and understanding, making it suitable for learning and experimentation. However, it does not take performance considerations into account and is not recommended for production applications. For production use cases, consider using libraries like TensorFlow.js instead.

Demos: https://neuroflow.andypai.me/

Installation

You can install NeuroFlow using npm:

npm install @andypai/neuroflow

Basic Usage

Here's an example of how you can train a simple neural network using the Layer and Sequential classes for a regression problem:

import { Sequential, Layer } from 'neuroflow'

// Specify the architecture
const layer1 = new Layer({ numOfInputs: 3, numOfNeurons: 4 })
const layer2 = new Layer({ numOfInputs: 4, numOfNeurons: 3 })
const layer3 = new Layer({
  numOfInputs: 3,
  numOfNeurons: 1,
  activation: 'linear',
})
const model = new Sequential({
  layers: [layer1, layer2, layer3],
})

// Training data
const xs = [
  [2.0, 3.0, -1.0],
  [3.0, -1.0, 0.5],
  [0.5, 1.0, 1.0],
  [1.0, 1.0, -1.0],
]
const ys = [1.0, -1.0, -1.0, 1.0]

// Training loop
const epochs = 15
for (let epoch = 0; epoch < epochs; epoch++) {
  // Forward pass
  const yPredictions = xs.map((x) => model.forward(x))

  // Mean squared error (MSE) loss
  const loss = ys.reduce(
    (acc, yTarget, i) => yPredictions[i].sub(yTarget).pow(2).add(acc),
    0,
  )

  // Backward pass
  model.zeroGrad()
  loss.backward()

  // Update weights
  const learningRate = 0.01
  model.parameters().forEach((p) => {
    p.data -= learningRate * p.grad
  })

  console.info(`Epoch: ${epoch + 1}, Loss: ${loss.data}`)
}

// Inference
model.forward([2.0, 3.0, -1.0]) // Output: 1

The full example is available here

Examples

License

This project is licensed under the MIT License.