Skip to content

Latest commit

 

History

History
88 lines (56 loc) · 3.59 KB

INSTALLING_PYTHON.md

File metadata and controls

88 lines (56 loc) · 3.59 KB
Raw

Installing Python Environment

In order to run the Jupyter notebooks locally, you need to have Python 3.8 installed. To check if you already have it, open up a terminal and run

  • Windows: run py --list-paths and check if 3.8 is in the list
  • macOS/Linux: run which python3.8 and check if the executable is found

If you have it installed, you can skip the next section and continue with Install Pipenv.

Install Python 3.8

Windows

Download and run the installer from https://www.python.org/ftp/python/3.8.6/python-3.8.6-amd64.exe

Make sure you select the Add Python 3.8 to PATH option during installation:

Note: this will prepend the Python 3.8 paths to your PATH, i.e. this will essentially become the default Python version inside your shell. For example, when you run a script from the command line with python my_script.py, the directories in PATH will be searched for the python executable and the first match is used.

In case you have a different Python version installed and depend on it being the default, you might want to change the order of your PATH entries (Windows Key > Edit the System Environment Variables > Environment Variables... > Edit your User Path variable) to move the Python 3.8 entries below your preferred default version, e.g.:

macOS

Just download and run the installer from https://www.python.org/ftp/python/3.8.6/python-3.8.6-macosx10.9.pkg

Linux

  • Use your distribution's package manager to install Python 3.8. Common package names are e.g. python3.8 or python38
  • Run python3.8 -m ensurepip --user to ensure you have the corresponding pip version installed

Install pipenv

In order not to break any of your existing Python installations, we will be creating a virtual environment for this tutorial using a tool called pipenv. It is just a Python package itself, so you can install it into your Python 3.8 distribution by running

pip3.8 install pipenv

in a terminal.

Note for Windows: if this command does not work and you already had Python 3.8 installed, it's probably missing from your PATH. Find out the installation directory with py --list-paths and add the two paths like in the "Edit environment variable" image above to your PATH environment variable.

Working with pipenv

Creating a New Virtual Environment

In the pipenv world, virtual environments are project-based. Each project needs to have special text file within the project directory. This so-called Pipfile contains the information about the environment (required Python version, dependencies, ...).

To create a new virtual environment for a project containing a Pipfile, you just need to do the following steps:

  • open a terminal
  • change into the root directory of the project (i.e. the directory containing the Pipfile)
  • run pipenv install

This will automatically create the new virtual environment using the specified Python version and install all required package dependencies. Depending on the number of packages to install this might take a while.

Using the Virtual Environment

Once the environment has been successfully created, there are basically two ways to use it:

  • pipenv run: run a single given command inside the virtual environment, e.g. starting the Python interpreter by running pipenv run python. All command arguments will be forwarded.
  • pipenv shell: spawn a subshell with the virtual environment activated (type exit to deactivate it again)

Note that these commands should always be executed from within the project directory!