Digital Education Resources - Vanderbilt Libraries Digital Lab

Previous lesson: Programming environments

Setting up VS Code to use Jupyter notebooks and GitHub Copilot

GitHub Copilot is an artificial intelligence autocompletion code suggestion tool. It is based on an OpenAI system similar to ChatGPT, but trained on code that has been submitted to GitHub. In this lesson, we will install Visual Studio Code (VS Code) and enable GitHub Copilot for use within the VS Code environment. Copilot is free for students and teachers. For others, it has a 30-day trial followed by a paid subscription.

Learning objectives At the end of this lesson, the learner will:

Example Jupyter notebook at GitHub

Getting access to GitHub Copilot

GitHub Copilot is a paid product, but it is free for students and teachers. If you are not a student or teacher, you can get a 30-day free trial. After that, you will have to pay for a subscription.

Because there is a waiting time for approval for the free educational access, you should start the process as soon as possible, since you may have to wait for up to several days for your application to be processed.

The first step in activating GitHub Copilot is to create a GitHub account if you don’t already have one. They are free and you can sign up at https://github.com/. Sign into your account and make note of the username you chose. Complete your Public Profile and make sure that you give your name exactly like it will appear on your academic ID. Also, complete the Bio section.

NOTE: when you sign up, if you do not use your school email you will need to add it as an additional email and verify it. You can do this under the settings, email. After adding the email, verify using the email message that is generated and sent to you. However, it is simplest to just sign up for GitHub using your school email if you plan to get the free educational access.

Obtaining free educational access

Once you have the account you need to sign up for free access if you have a school email address. Note that for either student or teacher free access, you have to submit a photo of your academic ID.

If you are a student:

  1. Go to https://education.github.com/. Under the Student menu, select Student Developer Pack.
  2. Click the Yes, I'm a student button.
  3. Select “Student” radio button if it isn’t already selected.
  4. Select school address. If it is not already there, go to github account settings, email, (click the button on the page) and verify.
  5. Fill in the section on how you will use GitHub.
  6. On next page, you need to submit a photo of your student ID.
  7. If all goes well and your application processes successfully, you will either get an email within an hour or within 10 days.

If you are a teacher:

Follow the instructions on the Apply to GitHub Global Campus as a teacher page. You may have to wait a few days for your application to be reviewed.

Installation

There are various ways to install VS Code and run Jupyter notebooks in the VS Code environment. You can find them by searching the web and reading about them.

In these instructions, we will consider two options: installing a Jupiter environment as part of the Anaconda distribution, and installing Jupyter notebooks directly. Installing via Anaconda is probably more straightforward and if you already have Anaconda installed, the setup is easier. However, it is fairly frequent for people to have problems with the Anaconda installation, so a direct install may be a better option in that situation.

What is Anaconda? (6m30s)

Anaconda is an umbrella system for data science that includes many of the most important tools used in data science. It is also free. It includes both the Python and R programming languages, most of the common Python libraries used in science and engineering (including NumPy, SciPy, Matplotlib, and pandas), and many commonly used R packages. Anaconda also includes the popular Jupyter notebook system, RStudio, the Spyder Python development environment, and has its own custom package management system. The Anaconda Navigator provides access to the system through a desktop GUI (shown in the screenshot above).

Technically, there are two different pieces in play. Anaconda itself is a software distribution - it includes a number of pre-configured programs and packages. Conda is a package manager that is installed automatically when Anaconda is installed. Conda is used to install, remove, and update the packages associated with the software that is a part of the Anaconda distribution. Conda can actually be used as a package management system independently of Anaconda. This blog post contains more details.

Is Anaconda for me?

Using Anaconda is appealing because it allows you to have access to many data science tools with a single install. However, there are several things to consider before installing Anaconda.

  1. It’s big. Because Anaconda automatically installs a lot of packages, it can get big (can be up to 3 GB). You might also have problems if your computer is using an outdated operating system. So Anaconda could be a problem on old computers with limited drive space.

  2. Potential conflicts. There could be potential conflicts between the Conda package managing system and other systems you are using. In particular, if you are using a Mac with Homebrew, you should do some further research before installing Anaconda. Here’s a place to start.

  3. Virtual environments. The Conda package manager has its own built-in environment manager. (That’s why a Mac Terminal prompt starts including (base) after Anaconda is installed.) There may be some complications if you need to use Conda-managed packages within your virtualenv or venv virtual environment. See this page for more details.

If you are a newby user with a reasonably new computer, you are probably not going to know or care about any of these things and can probably safely just install Anaconda and get on with your life. If you are an advanced user and completely understand all of these things, then you will understand the implacations and be able to deal with any problems that may arise. Things might be complicated if you are an intermediate user and are using the tools mentioned above (e.g. Homebrew or virtual environments), but aren’t expert enough to figure out how to fix things if they go wrong.

Installing VS Code with Jupyter installed via Anaconda Navigator (Option 1)

If you decide that Anaconda is for you, go to the Anaconda Installation page and folow the links for your operating system.

The initial installation of VS Code will need to be done separately. To install VS Code, go to the downloads page and follow the instructions for your operating system.

To start using Anaconda, go to the Anaconda Navigator. It will show up in your Windows Start menu or Launchpad on Mac. Once you have installed VS Code, it should show up in the Anaconda Navigator, from which you can lauch it. Since Anaconda includes a Jupyter environment, there are no additional steps to install Jupyter. The Anaconda Getting started page has more details.

Installing VS Code and a Jupyter environment directly (Option 2)

To install VS Code, go to the downloads page and follow the instructions for your operating system.

The easiest way to install a Jupyter environment is to use the Python package manager, PIP. In order to do that, you need to have Python 3 installed on your computer. Depending you your system, you may already have it. To check, open a terminal window and type:

python3 --version

If you get a version number, you have Python 3 installed. If not, try typing:

python --version

If you get a Python 3 version number, you have Python 3 installed. If you get a Python 2 version number or an error message, you can download Python 3 from the Python downloads page. For detailed installation instructions, see this page.

Once you have determined that you have Python3, in the terminal window type:

pip3 install jupyter

or

pip install jupyter

depending on your system. This will install the Jupyter environment.

Installing the Python and Jupyter extensions in VS Code

Once you have installed VS Code, you need to install the Python and Jupyter extensions. To do this, open VS Code and click on the Extensions icon in the left sidebar. In the search box, type Python and click on the Install button for the Python extension. Then type Jupyter and click on the Install button for the Jupyter extension.

Activating an environment in a Jupyter notebook

There are two aspects to running a Jupiter notebook in VS Code. One is manipulating the notebook cells and writing code in them, and the other is getting the notebook to actually run in an environment. In theory, you should activate an Anaconda environment before you run the notebook, but you can get away with just editing the notebook and the first time you try to run a cell, you will be prompted to activate an environment.

To activate the environment first, open the Command Pallette by pressing the shift and command then P keys on a Mac. On Windows, press Shift and Ctrl then P. From the menu, select Python: Select Interpreter. This will bring up a popup from which you can select the installation of Python you want to use. Generally, there will be a Recommended one and it is usually safe to just select that. The main reason this makes a difference is that you may have different packages installed in different environments. For example, if you select the Anaconda installation, it will probably have many of the typical libraries you would want to use already installed. If you pick a generic installation, you may need to install some packages from the command line using PIP.

If you already have a notebook open, you’ll see a Select Kernal option in the upper right. Clicking on that will bring up the same options that you’ll see in the Python: Select Interpreter dialog. Similarly, if you run a cell without setting an environment, the selection dialog will pop up at the top of the screen after you click the run button.

Using Jupyter notebooks in VS Code

Downloading and running a Jupyter notebook from GitHub (1m54s)

Example Jupyter notebook to download from GitHub

The video above shows how to run the downloaded notebook from the Jupyter notebook web interface. However, we will be opening the notebooks using VS Code. From within VS Code, go to the file menu and select Open.... Navigate to the notebook you want to open and select it. The notebook will open in a new tab.

If you want notebooks to open in VS Code by default when you double-click on a notebook icon in your file system navigator, you can set that using these instructions for Mac or these instructions for Windows. The file extension for Jupyter notebooks is .ipynb.

Workspace Trust

If you work with a new notebook that you’ve downloaded from somewhere else, you can look at the code but you can’t run it without indicating that you trust it. There will be a popup dialog at the top of the screen for you to do this. If this annoys you, you can set a particular folder to be trusted by default. All notebooks run from that folder will be trusted and you won’t have to go through the dialog each time.

Running cells

To run a cell, click on the “play” button at the left of the cell. Any output will appear below the cell. There are buttons at the top of the screen to Clear All Outputs and Restart (which clears the values of all variables in the environment). These two functions are independent – clearing the output does not clear the environment and vice versa.

Using GitHub Copilot in the VS Code environment

Activating GitHub Copilot

Instructions based on this detailed page

  1. Go to https://github.com/github-copilot/free_signup. You should see “Congratulations! You are eligible to use GitHub Copilot for free.” Click on the Get access to GitHub Copilot button.
  2. On the “Select your preferences” page, select “Allow”. Chose whether you want to allow GitHub to use your code for training (optional). Click the Save and get started button.
  3. In the Visual Studio Code Marketplace, go to the GitHub Copilot extension page and click Install.
  4. A popup will appear, asking to open Visual Studio Code. Click Open Visual Studio Code.
  5. When VS Code opens, you will be in a “Extension: GitHub Copilot” tab. There should be an Install button. Click it to complete the install.
  6. You should get a prompt to authorize GitHub for VS Code. Click the Authorize Visual-Studio-Code button. If you don’t see this prompt, close and reopen VS Code. Allow it to open the link.
  7. In the lower right corner you should see the Copilot icon. It looks like this:

Copilot icon

If not, restart VS Code.

Using GitHub Copilot

GitHub Copilot is an autocompletion tool, meaning that tt will suggest code to you as you type. We will be using it to help us write Python code, but it can also be used for other languages such as JavaScript. It can actually be used with Markdown as well, and made suggestions for writing the text on this page.

When using GitHub Copilot in a Jupyter notebook, the easiest thing to do is to type a comment in a cell (comments begin with tht # character). Copilot will then suggest code that it thinks will do what you want. You can then select the code you want and press the Tab key to insert it into the cell. If the suggested code is not what you want, you can start typing and Copilot will make new suggestions. Usually it will suggest a single line of code, but sometimes it will suggest an entire block of code or function if it thinks that is what you want.

Next Python lesson: Python programming basics

Revised 2023-10-05

Questions? Contact us

License: CC BY 4.0.
Credit: "Vanderbilt Libraries Digital Lab - www.library.vanderbilt.edu"