Getting Started with Colab

Goals of this page #

• Understand the basics of Google Colab and run your first Notebook
• Learn the advantages and caveats of working in Colab
• Collect tips for migrating to a local environment later

Why choose Colab? #

If you are unsure about setting up locally, get comfortable with Python in Colab first, then step up to a uv-based environment later.

What you need #

1. Google account
   Create one at https://accounts.google.com/ if you do not already have it.
2. Browser
   Google Chrome (latest version) is recommended. Other Chromium-based browsers generally work too.

Create and run a Notebook #

1. Go to Google Colab and sign in with your Google account
2. Click “New notebook”
3. In the first cell, enter the following code and press Shift + Enter

print("Hello, Python from Colab!")

If you see a check mark and execution time on the left of the cell, it worked.

Working with files #

• Temporary files
  • Use the cell magic %%writefile sample.py to create a short-lived file.
  • Remember: when the runtime resets, temporary files are removed. Mount Drive if you need persistence.
• Mounting Google Drive

  from google.colab import drive
  drive.mount('/content/drive')
  

  • Follow the authorisation flow and agree; your Drive will appear under /content/drive/MyDrive.
• Upload / download files

  from google.colab import files
  files.upload()    # opens a file chooser dialog
  files.download("result.csv")
  

Runtime types #

• Menu “Runtime” → “Change runtime type” lets you switch to GPU or TPU
• The free tier has time limits, so it is not suitable for very long jobs
• Features that keep the session alive automatically can violate the terms—check the usage policy

Installing common libraries #

Colab comes with many libraries pre-installed, but you can pin versions with pip.

!pip install pandas==2.2.1
!pip install "scikit-learn>=1.4,<1.5"

Prefix ! to execute shell commands. Note that you need to reinstall packages each time the runtime restarts.

Organising notebooks #

• Notebooks are saved to Google Drive (default: My Drive/Colab Notebooks)
• Use “File” → “Save a copy in Drive” to clean up folders and keep things tidy
• If you need versioning/history, connect the notebook to GitHub

Limitations and workarounds #

Tips for moving to a local environment #

• Download notebooks (.ipynb) and open them in VS Code to keep working locally
• Export package versions to requirements.txt so you can recreate the setup later
• Once dependencies grow, manage them with commands such as uv pip freeze > requirements.txt

For learning syntax and control flow, Colab is more than enough. When you are ready to try command-line work or virtual environments, continue with the local setup guides:

• Set up uv on Windows
• Set up uv on macOS
• Set up uv on Ubuntu

When you’re ready, move on to the next lesson to start learning Python.