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.
TensorFlow.js for Node currently supports the following platforms:
- Mac OS X CPU (10.12.6 Siera or higher)
- Linux CPU (Ubuntu 14.04 or higher)
- Linux GPU (Ubuntu 14.04 or higher and Cuda 9.0 w/ CUDNN v7) (see installation instructions)
- Windows CPU (Win 7 or higher)
- Windows GPU (Win 7 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-nodeyarn add @tensorflow/tfjs-node
Installing Linux/Windows GPU TensorFlow.js for Node:
npm install @tensorflow/tfjs-node-gpuyarn add @tensorflow/tfjs-node-gpu
Windows Requires Python 2.7
Windows build support for
node-gyp requires Python 2.7. Be sure to have this version before installing
@tensorflow/tfjs-node-gpu. Machines with Python 3.x will not install the bindings properly.
For more troubleshooting on Windows, check out WINDOWS_TROUBLESHOOTING.md.
Mac OS X Requires Xcode
If you do not have Xcode setup on your machine, please run the following commands:
After that operation completes, re-run
yarn add or
npm install for the
You only need to include
@tensorflow/tfjs-node-gpu in the package.json file, since those packages ship with
Using the binding
Before executing any TensorFlow.js code, import the node package:
// Load the binding;// Or if running with GPU:;
Note: you do not need to add the
@tensorflow/tfjs package to your dependencies or import it directly.
# 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
MNIST demo for Node.js
See the tfjs-examples repository for training the MNIST dataset using the Node.js bindings.
Optional: Build libtensorflow From TensorFlow source
This requires installing bazel first.
bazel build --config=monolithic //tensorflow/tools/lib_package:libtensorflow