Installation#

WindKit is distributed via conda and PyPI. Due to dependencies on GIS utilities (GeoPandas, Rasterio) that require C libraries (GDAL, GEOS, PROJ), we recommend installing via conda for the smoothest experience.

Platform Support#

Platform

Status

Notes

Linux / WSL

Supported

Recommended platform

Windows

Supported

macOS

Supported

Prerequisites#

Before installing, ensure you have:

Package Manager

We recommend Miniforge (provides mamba) or Pixi for fast, reliable dependency resolution.

Python Version

WindKit supports Python 3.10 to 3.13.

Quick Install#

For experienced users, here are the essential commands:

mamba create -n windkit -c https://conda.windenergy.dtu.dk/channel/open/ -c conda-forge windkit
mamba activate windkit

pixi init && pixi project channel add https://conda.windenergy.dtu.dk/channel/open/
pixi add windkit

pip install windkit

Detailed Installation#

Step 1: Install a package manager

We recommend Miniforge, which provides both conda and the faster mamba.

Linux:

wget https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-Linux-x86_64.sh
bash Miniforge3-Linux-x86_64.sh

Follow the prompts and restart your terminal.

Note

Windows users

Download and run the latest Windows installer from Miniforge releases. Use the default options, then open Miniforge Prompt from the Start menu. To use PowerShell instead, run conda init powershell and restart PowerShell.

Step 2: Create environment and install WindKit

mamba create -n windkit -c https://conda.windenergy.dtu.dk/channel/open/ -c conda-forge windkit

Step 3: Activate and verify

mamba activate windkit
python -c "import windkit; print(windkit.__version__)"

Pixi is a modern, fast package manager for project-based dependency management.

Step 1: Install pixi

Follow the instructions at pixi.sh.

Step 2: Create a project

mkdir my-windkit-project
cd my-windkit-project
pixi init

Step 3: Add channels and install WindKit

pixi project channel add https://conda.windenergy.dtu.dk/channel/open/
pixi add windkit

Step 4: Verify installation

pixi run python -c "import windkit; print(windkit.__version__)"

To work interactively, use pixi shell to enter the environment.

Warning

PyPI installation requires you to install C library dependencies (GDAL, GEOS, PROJ) yourself. We recommend conda for easier setup.

Using pip:

pip install windkit

Using uv (faster alternative):

uv venv && source .venv/bin/activate
uv pip install windkit

With optional dependencies:

pip install windkit[plots,data_apis,performance]

Post-Installation Setup#

Configure WindKit (optional)

WindKit can be configured via a TOML file or environment variables. See the Configuration.

Save channel credentials (optional, conda/mamba only)

To avoid typing channel URLs for future installs:

mamba config --add channels https://conda.windenergy.dtu.dk/channel/open/
mamba config --add channels conda-forge

Updating WindKit#

We recommend creating a new environment for major updates:

mamba create -n windkit_new -c https://conda.windenergy.dtu.dk/channel/open/ -c conda-forge windkit

pixi update

pip install --upgrade windkit

Troubleshooting#

Import errors

ModuleNotFoundError: No module named ‘windkit’

Ensure your environment is activated:

mamba activate windkit  # or: pixi shell

ImportError: DLL load failed (Windows)

Try reinstalling in a fresh environment:

mamba create -n windkit_fresh -c https://conda.windenergy.dtu.dk/channel/open/ -c conda-forge windkit
GDAL/GEOS/PROJ errors (pip installs)

If you installed via pip and see errors about missing GDAL, GEOS, or PROJ:

  1. Install system dependencies. See the installation guides for GeoPandas and Rasterio.

  2. Or switch to conda installation (recommended).

Dependencies#

Required dependencies
Optional dependencies

Plotting (windkit[plots]):

Data APIs (windkit[data_apis]):

Performance (windkit[performance]):

  • numba (0.57+) - JIT compilation for faster functions

Need Help?#

If you encounter issues not covered here:

See also