Installation

Requirements

  • Python ≥ 3.10
  • Recommended: virtual environment

Bulldozer depends on scientific Python libraries such as numpy, scipy, and rasterio.
On some systems (especially clusters), rasterio may require system-level GDAL libraries.

The simplest way to install Bulldozer is via pip:

python -m venv venv
source venv/bin/activate
pip install bulldozer-dtm

Then verify installation:

bulldozer --help

Install from source (GitHub)

Repository: https://github.com/CNES/bulldozer

git clone https://github.com/CNES/bulldozer.git
cd bulldozer

make install
source bulldozer_venv/bin/activate

This will: - Create a dedicated virtual environment (bulldozer_venv) - Install all runtime and development dependencies - Install pre-commit hooks

Manual installation with venv

git clone https://github.com/CNES/bulldozer.git
cd bulldozer

python -m venv bulldozer_venv
source bulldozer_venv/bin/activate
pip install .

Development installation

To install Bulldozer in editable mode with development tools:

pip install -e .[dev]

This installs:

  • Ruff (lint & format)
  • mypy
  • pytest
  • pre-commit

Using Docker

A Docker image is available on Docker Hub:

docker pull cnes/bulldozer:latest

Example usage:

docker run --rm -v $(pwd):/data cnes/bulldozer:latest bulldozer --help