Counting 2,784 Big Data & Machine Learning Frameworks, Toolsets, and Examples...
Suggestion? Feedback? Tweet @stkim1

Author
Contributors
Last Commit
May. 27, 2018
Created
Feb. 12, 2018

TensorFlow backend for TensorFlow.js via Node.js

This repo is under active development and is not production-ready. We are actively developing as an open source project.

Installing

TensorFlow.js for Node currently supports the following platforms:

  • Mac OS X 10.12.6 (Siera) or higher
  • Linux CPU (Ubuntu 16.04 or higher)
  • Linux GPU (Ubuntu 16.04 or higher and Cuda 9.0 w/ CUDNN v7) (see installation instructions)

Other Linux variants might also work but this project matches core TensorFlow installation requirements.

Installing CPU TensorFlow.js for Node:

npm install @tensorflow/tfjs-node
(or)
yarn add @tensorflow/tfjs-node

Installing Linux GPU TensorFlow.js for Node:

npm install @tensorflow/tfjs-node-gpu
(or)
yarn add @tensorflow/tfjs-node-gpu

Before executing any TensorFlow.js code, load and set the backend to 'tensorflow'.

import * as tf from '@tensorflow/tfjs';

// Load the binding
import '@tensorflow/tfjs-node';

// Or if running with GPU:
import '@tensorflow/tfjs-node-gpu';

tf.setBackend('tensorflow');

Development

# Download and install JS dependencies, including libtensorflow 1.8.
yarn

# Run TFJS tests against Node.js backend:
yarn test
# Switch to GPU for local development:
yarn enable-gpu

See the demo directory that trains MNIST using TensorFlow.js with the TensorFlow C backend.

cd demo/
yarn

# Run the training script. See demo/package.json for this script.
yarn mnist

The important line to note is at the top of mnist.ts, which sets the backend to TensorFlow.

Optional: Build libtensorflow From TensorFlow source

This requires installing bazel first.

bazel build --config=monolithic //tensorflow/tools/lib_package:libtensorflow

Latest Releases
v0.1.5
 May. 24 2018
v0.1.4
 May. 23 2018
v0.1.3
 May. 22 2018
v0.1.2
 May. 8 2018
v0.1.1
 May. 8 2018