.. include:: ../color_roles.rst Installation ======================== Fragile has been tested in Ubuntu 18.04 and Ubuntu 19.04. It supports Python 3.6, 3.7 and 3.8. If you find any problems running it in a different OS or Python version please `open an issue `_. Installing from pip ^^^^^^^^^^^^^^^^^^^^^ Fragile can be installed from ``pip`` running .. code-block:: bash pip install fragile This will install the package and the dependencies for using the ``core`` and ``optimize`` modules. It is also possible to install with pip the dependencies for using the other modules You can install the dependencies for running Atari games with: ``pip install fragile["atari"]`` The dependencies for the ``dataviz`` module can be installed running: ``pip install fragile["dataviz"]`` To install the dependencies for the ``distributed`` module you can run: ``pip install fragile["ray"]`` The dependencies for running the tests can be installed with: ``pip install fragile["test"]`` All the dependencies can be installed at once running: .. code-block:: bash pip install fragile["all"] Installing from source ^^^^^^^^^^^^^^^^^^^^^^^^ To install ``fragile`` from source you can run: .. code-block:: bash git clone https://github.com/FragileTech/fragile.git cd fragile pip3 install -r requirements.txt pip3 install -r requirements-viz.txt pip3 install -e .["all"] Running in Docker ^^^^^^^^^^^^^^^^^^ The fragile docker container will execute a Jupyter notebook accessible on port 8080 with password: ``fragile`` Pulling the Docker image from Docker Hub ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ You can pull a docker image from Docker Hub running: .. code-block:: bash docker pull fragiletech/fragile:version-tag Where version-tag corresponds to the fragile version that will be installed in the pulled image. Building the Docker image from source ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ After cloning the repository run: .. code-block:: bash make docker build docker run -d -p 8080:8080 -v PATH_TO_REPO/fragile fragile You can also run the tests inside the docker container .. code-block:: bash make docker-test You can change the default jupyter notebook password by passing ``JUPYTER_PASSWORD=your_password`` as a docker build argument.