Instructions to create and intsall TensorFlow Quantum as a PIP package:

First create a Docker container with Ubuntu 16.04 and the devtoolset7 toolchain.

./release/open_ubuntu_docker.sh

Then build pip packages for python 3.6 and 3.7 inside of the docker with:

./release/build_all_wheels.sh
exit

The resulting .whl files will be placed in a new folder outside of the docker called wheels. From here they need to be repaired with auditwheel to ensure maximum compatability across platforms. To repair the wheels run:

./release/repair_wheels.sh

Now the wheels folder should contain the built wheel files and the manylinux2010 version of the wheels for python 3.6 and 3.7.

A wheel can be installed with:

python3 -m pip install --user wheels/name_of_wheel.whl

Note, that if you are planning on running TFQ as a PIP package instead of with Bazel, you cannot run from the tensorflow_quantum/ directory, or python will attempt to use the local files instead of the site-package/ files and fail to find dependencies.