Installing Python Packages with pip and PyPI Guide

Learn how to use pip to install, update, and remove Python packages from PyPI, manage requirements.txt, and avoid common installation pitfalls.

One of Python’s greatest strengths is the enormous ecosystem of open-source packages available for almost every task imaginable. Instead of writing complex code from scratch to download web data, process spreadsheets, or call an AI API, you can install a package written and maintained by other developers and use it in your project immediately. Understanding how to manage these packages with pip is one of the most practical skills you will use every day as a Python developer.

What are packages?

Packages are collections of Python code organised to solve specific problems. They are distributed through PyPI — the Python Package Index — which hosts over 500,000 packages.

Some packages you will use frequently in this course:

Package	What it does
requests	Sends HTTP requests to download web data and call APIs
pandas	Reads, cleans, and analyses tabular data
numpy	Fast numerical operations and array manipulation
openai	Connects to OpenAI’s AI models
python-dotenv	Loads environment variables from .env files
ipykernel	Enables interactive Python execution in VS Code

Meet pip

pip (Pip Installs Packages) is Python’s built-in package manager. It comes with Python and is already available inside your virtual environment. It handles downloading packages from PyPI, installing them, resolving their dependencies, and managing versions.

When you activate your virtual environment, you get a dedicated copy of pip that installs packages only into that environment. Packages installed here do not affect any other project or your system Python.

Install your first package

1. Open the VS Code terminal

   Press Ctrl + ` (Windows/Linux) or Cmd + ` (macOS). Confirm you see (.venv) at the start of the prompt — this means your virtual environment is active.

2. Install a package

   pip install requests
   

   You will see output like:

   Collecting requests
     Downloading requests-2.31.0-py3-none-any.whl (62 kB)
   Installing collected packages: requests
   Successfully installed requests-2.31.0
   
3. Use it in your code

   import requests
   
   response = requests.get("https://api.github.com")
   print(response.status_code)  # 200 means success
   

Notice that installing requests also installed packages like certifi, charset-normalizer, and urllib3. These are dependencies — packages that requests itself relies on. pip resolves and installs the full dependency tree automatically.

To see the actual files that were installed, look inside .venv/lib/python3.x/site-packages/requests/ — it is just Python files that someone else wrote and shared.

Common pip commands

# Install a package
pip install requests

# Install multiple packages at once
pip install requests pandas numpy

# Install a specific version
pip install requests==2.30.0

# Install minimum version
pip install requests>=2.28.0

# Upgrade an installed package
pip install --upgrade requests

# Uninstall a package
pip uninstall requests

# List all installed packages
pip list

# Show details about a specific package
pip show requests

Managing dependencies with requirements.txt

When you share your project with others (or deploy it to a server), they need to know which packages to install and which versions you used. A requirements.txt file captures this information.

Generate a requirements file:

pip freeze > requirements.txt

The file looks like this:

certifi==2024.2.2
charset-normalizer==3.3.2
idna==3.6
requests==2.31.0
urllib3==2.2.0

Install from a requirements file:

pip install -r requirements.txt

This is the standard way to reproduce an environment — on a new machine, in a team, or in a CI/CD pipeline. Anyone with your requirements.txt can recreate an identical environment with a single command.

Commit your requirements.txt to version control (Git) so your teammates and future deployments always have a record of the exact package versions your project depends on.

Troubleshooting

pip: command not found

Your virtual environment is probably not active. Check whether (.venv) appears in your terminal prompt. If not, select your interpreter in VS Code via Ctrl/Cmd + Shift + P → Python: Select Interpreter.

Permission denied

You are probably trying to install into system Python rather than your virtual environment. Never use sudo pip install. Always confirm (.venv) is visible in your prompt before installing.

Package not found

Package names on PyPI can differ from what you might expect:

• “Beautiful Soup” → pip install beautifulsoup4
• “OpenAI” → pip install openai
• “Pillow (PIL)” → pip install pillow

Search pypi.org for the exact package name if you are unsure.

Version conflict errors

If pip reports a conflict between package versions, the cleanest solution is to create a fresh virtual environment and reinstall only what your project needs. This is exactly why you use one virtual environment per project.

Interactive Python

Learn how to run Python interactively from a .py file using Shift + Enter.