This guide will help you set up a Python virtual environment for your projects, using both the standard `venv`/`pip` method and the modern `uv` tool. These instructions are focused on Windows, but notes for macOS/Linux are included where relevant.
A virtual environment allows to:
- Isolate project dependencies
- Avoid conflicts between package versions
- Reproduce environments easily for development, testing, and deployment
Prerequisites
- Python installed (recommended: 3.8+ for stability, but 3.12+ is supported)
- (Optional) Multiple Python versions can be managed by installing them in separate directories
- Below setup will focus Windows system. And commands to setup may vary for macOS or Linux
- For macOS/Linux, replace backslashes (`\`) with slashes (`/`) and use `source /bin/activate` to activate
- Always activate your virtual environment before running or installing packages
- For projects using Jupyter, install the kernel inside the virtual environment for isolation
- Use `uv` for faster installs and modern dependency management, especially for larger projects
Setup
1. Create a Virtual Environment
# Replace <python_path> with your Python executable path
# Example for custom Python install:
C:\Users\<username>\pyver\py3121\python -m venv <env_name>
2. Activate the Virtual Environment
# In PowerShell or Command Prompt:
<env_name>\Scripts\activate
3. Install Project Dependencies
pip install -r requirements.txt
4. (Optional) Jupyter Notebook Setup
pip install jupyter ipython ipykernel
python -m ipykernel install --user --name=<env_name>
# To remove a kernel:
jupyter kernelspec uninstall <env_name>
5. (Optional) Install Bash Kernel for Jupyter
pip install bash_kernel
python -m bash_kernel.installuv is a fast Python package/dependency manager and virtual environment tool.
Setup
1. Install uv (globally)
pip install uv
2. Create a Virtual Environment
uv venv <folder_name>
# This creates a .venv directory inside <folder_name>
3. Activate the Virtual Environment
<folder_name>\.venv\Scripts\activate
4. Add or Install Packages
# To add a package:
uv add <package-name>
# To install from requirements.txt:
uv pip install -r requirements.txt
# To install from pyproject.toml/uv.lock:
uv pip install -e .
5. Export and Lock Dependencies
# Export a requirements.txt from uv.lock:
uv export --format requirements-txt
# Create or update a uv.lock file:
uv lock
# Add dependencies from requirements.txt and update uv.lock:
uv add -r requirements.txt
6. Project Initialization and Advanced uv Usage
# To initialize a new project (creates pyproject.toml, uv.lock, etc.):
uv init <name-of-project>
# To install packages from requirements.txt:
uv pip install -r requirements.txt
# To install the current project in editable mode:
uv pip install -e .
# To create or update a uv.lock file:
uv lock
# To add dependencies from requirements.txt to pyproject.toml and create/update uv.lock:
uv add -r requirements.txt