install.md

Install TensorFlow Federated

There are a few ways to set up your environment to use TensorFlow Federated (TFF):

• The easiest way to learn and use TFF requires no installation—run the TensorFlow Federated tutorials directly in your browser using Google Colaboratory.
• To use TensorFlow Federated on a local machine, install the TFF package with Python's pip package manager.
• If you have a unique machine configuration, build TensorFlow Federated from source.

Install TensorFlow Federated using pip

1. Install the Python development environment.

On Ubuntu:

sudo apt update
sudo apt install python3-dev python3-pip  # Python 3
sudo pip3 install --upgrade virtualenv  # system-wide install

On macOS:

/usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"
export PATH="/usr/local/bin:/usr/local/sbin:$PATH"
brew update
brew install python  # Python 3
sudo pip3 install --upgrade virtualenv  # system-wide install

2. Create a virtual environment.

virtualenv --python python3 "venv"
source "venv/bin/activate"
pip install --upgrade pip

Note: To exit the virtual environment, run deactivate.

3. Install the TensorFlow Federated pip package.

pip install --upgrade tensorflow_federated

4. (Optional) Test Tensorflow Federated.

python -c "import tensorflow_federated as tff; tff.federated_computation(lambda: 'Hello, World!')()"

Success: TensorFlow Federated is now installed.

Build the TensorFlow Federated pip package

Build the TensorFlow Federated pip package and install it on Ubuntu or macOS.

1. Create a development environment using virtualenv or using Docker.
2. Build the pip package.

Using virtualenv

Create a Tensorflow Federated development environment using virtualenv on Ubuntu or macOS.

1. Install the Python development environment.

On Ubuntu:

sudo apt update
sudo apt install python3-dev python3-pip  # Python 3
sudo pip3 install --upgrade virtualenv  # system-wide install

On macOS:

/usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"
export PATH="/usr/local/bin:/usr/local/sbin:$PATH"
brew update
brew install python  # Python 3
sudo pip3 install --upgrade virtualenv  # system-wide install

2. Install Bazel.

Install Bazel, the build tool used to compile Tensorflow Federated.

Note: Bazel version 0.19.2 or greater is required by TensorFlow Federated.

3. Clone the Tensorflow Federated repository.

git clone https://github.com/tensorflow/federated.git
cd "federated"

4. Create a virtual environment.

virtualenv --python python3 "venv"
source "venv/bin/activate"
pip install --upgrade pip

Note: To exit the virtual environment, run deactivate.

5. Install Tensorflow Federated dependencies.

pip install --requirement "requirements.txt"

6. (Optional) Test Tensorflow Federated.

bazel test //tensorflow_federated/...

Success: The TensorFlow Federated development environment is ready, now build the pip package.

Using Docker

Create a Tensorflow Federated development environment using Docker on Ubuntu or macOS.

1. Install Docker.

Install Docker on your local machine.

2. Clone the latest Tensorflow Federated source.

git clone https://github.com/tensorflow/federated.git
cd "federated"

3. Build a Docker image.

docker build . \
    --tag tensorflow_federated:latest

4. Start a Docker container.

docker run -it \
    --workdir /federated \
    --volume $(pwd):/federated \
    tensorflow_federated:latest \
    bash

5. (Optional) Test Tensorflow Federated.

bazel test //tensorflow_federated/...

Success: The TensorFlow Federated development environment is ready, now build the pip package.

Build the pip package

Build the TensorFlow Federated pip package and install it on Ubuntu or macOS.

1. Build the pip package.

mkdir "/tmp/tensorflow_federated"
bazel build //tensorflow_federated/tools:build_pip_package
bazel-bin/tensorflow_federated/tools/build_pip_package "/tmp/tensorflow_federated"

2. Create a new project.

mkdir "/tmp/project"
cd "/tmp/project"

virtualenv --python python3 "venv"
source "venv/bin/activate"
pip install --upgrade pip

Note: To exit the virtual environment run deactivate.

3. Install the pip package.

pip install --upgrade "/tmp/tensorflow_federated/tensorflow_federated-"*".whl"

4. Test Tensorflow Federated.

python -c "import tensorflow_federated as tff; tff.federated_computation(lambda: 'Hello, World!')()"

Success: The TensorFlow Federated package is built.