@tensorflow/tfjs-core

What is @tensorflow/tfjs-core?

@tensorflow/tfjs-core is a JavaScript library for developing and training machine learning models in the browser and on Node.js. It provides a flexible and efficient way to perform numerical computations using tensors, which are the core data structure for machine learning.

What are @tensorflow/tfjs-core's main functionalities?

Tensor Operations

This feature allows you to create and manipulate tensors, which are multi-dimensional arrays. The code sample demonstrates creating two tensors and performing an element-wise addition.

const tf = require('@tensorflow/tfjs-core');
const a = tf.tensor([1, 2, 3, 4]);
const b = tf.tensor([5, 6, 7, 8]);
const c = tf.add(a, b);
c.print();

Model Training

This feature allows you to define, compile, and train machine learning models. The code sample demonstrates creating a simple linear regression model, training it with some data, and making a prediction.

const tf = require('@tensorflow/tfjs-core');
const model = tf.sequential();
model.add(tf.layers.dense({units: 1, inputShape: [1]}));
model.compile({optimizer: 'sgd', loss: 'meanSquaredError'});
const xs = tf.tensor2d([1, 2, 3, 4], [4, 1]);
const ys = tf.tensor2d([1, 3, 5, 7], [4, 1]);
model.fit(xs, ys, {epochs: 10}).then(() => {
  model.predict(tf.tensor2d([5], [1, 1])).print();
});

GPU Acceleration

This feature leverages WebGL to perform computations on the GPU, which can significantly speed up numerical operations. The code sample demonstrates a simple tensor addition, which will be accelerated if a GPU is available.

const tf = require('@tensorflow/tfjs-core');
const a = tf.tensor([1, 2, 3, 4]);
const b = tf.tensor([5, 6, 7, 8]);
const c = tf.add(a, b);
c.print();

Other packages similar to @tensorflow/tfjs-core

onnxjs

ONNX.js is a JavaScript library for running ONNX (Open Neural Network Exchange) models in the browser and Node.js. It provides similar functionalities for running pre-trained models but focuses on the ONNX model format, which is widely used for interoperability between different machine learning frameworks.

brain.js

Brain.js is a JavaScript library for neural networks, which can be used in Node.js and the browser. It provides a simpler API compared to @tensorflow/tfjs-core and is more focused on ease of use for common neural network tasks, but it may not offer the same level of flexibility and performance.

README

TensorFlow.js Core API

NOTE: Building on the momentum of deeplearn.js, we have joined the TensorFlow family and we are starting a new ecosystem of libraries and tools for Machine Learning in Javascript, called TensorFlow.js. This repo moved from PAIR-code/deeplearnjs to tensorflow/tfjs-core.

A part of the TensorFlow.js ecosystem, this repo hosts @tensorflow/tfjs-core, the TensorFlow.js Core API, which provides low-level, hardware-accelerated linear algebra operations and an eager API for automatic differentiation.

Check out js.tensorflow.org for more information about the library, tutorials and API docs.

To keep track of issues, we use the tensorflow/tfjs Github repo.

Importing

You can install TensorFlow.js via yarn or npm. We recommend using the @tensorflow/tfjs npm package, which gives you both this Core API and the higher-level Layers API:

import * as tf from '@tensorflow/tfjs';
// You have the Core API: tf.matMul(), tf.softmax(), ...
// You also have Layers API: tf.model(), tf.layers.dense(), ...

On the other hand, if you care about the bundle size and you do not use the Layers API, you can import only the Core API:

import * as tfc from '@tensorflow/tfjs-core';
// You have the Core API: tfc.matMul(), tfc.softmax(), ...
// No Layers API.

For info about development, check out DEVELOPMENT.md.

For more information

• TensorFlow.js API documentation
• TensorFlow.js Tutorials

Thanks BrowserStack for providing testing support.