Attend the Women in ML Symposium on December 7 Register now

Install TensorFlow Ranking

Stay organized with collections Save and categorize content based on your preferences.

There are several ways to set up your environment to use the TensorFlow Ranking library.

  • The easiest way to learn and use TensorFlow Ranking is run any of the tutorials Google Colab. Select the link at the top of the Quickstart tutorial).
  • To use the Ranking library on a local machine, install the tensorflow_ranking pip package.
  • If you have a unique machine configuration, you can build the package from source, using the Build from source instructions.

Install TensorFlow Ranking using pip

Install using pip.

pip install --upgrade tensorflow_ranking

Build from source

You can also install from source, which requires the Bazel build system.

  1. Install Bazel, Git and Pip.

    sudo apt-get update
    sudo apt-get install bazel git python3-pip python3-venv
    
  2. Clone the TensorFlow Ranking repository.

    git clone https://github.com/tensorflow/ranking.git

  3. Build TensorFlow Ranking wheel file and store them in a /tmp/ranking_pip folder.

    cd ranking  # folder cloned in Step 2.
    bazel build //tensorflow_ranking/tools/pip_package:build_pip_package
    bazel-bin/tensorflow_ranking/tools/pip_package/build_pip_package \
        /tmp/ranking_pip
    
  4. Activate a venv environment.

    python3 -m venv --system-site-packages venv
    source venv/bin/activate
    
  5. Install the wheel package in your venv environment.

    pip install /tmp/ranking_pip/tensorflow_ranking*.whl
    
  6. Optionally, run all TensorFlow Ranking tests.

    bazel test //tensorflow_ranking/...
    

For more information about installing Python, pip, TensorFlow, and working with Python virtual environments, see Install TensorFlow with pip.