Install¶
Installing in an existing python environment with pip¶
First, make sure that numpy is installed in your python environment:
pip install 'numpy<2.0'
Then, install pNbody running:
pip install ./pNbody
where ./pNbody is the directory cloned from gitlab.
Installing pNbody in a specific directory is achieved with the
--prefix
option:
pip install --prefix=/where/you/want/to/install/it ./pNbody
Installing using a fresh Python Virtual Environment¶
Make sure that the PYTHONPATH variable is not set:
unset PYTHONPATH
Create a python virtual environment and activate it:
python3 -m venv ~local/python-venv/pNbody
source ~local/python-venv/pNbody/bin/activate
Install numpy:
pip install 'numpy<2.0'
Install pNbody:
pip install ./pNbody
Installing using a Conda Environment¶
Make sure that the PYTHONPATH variable is not set:
unset PYTHONPATH
Create a new conda environment and activate it:
conda create -n pNbody-env
conda activate pNbody-env
Install numpy:
conda install 'numpy<2.0'
Install pNbody:
pip install ./pNbody
Troubleshooting¶
In some rare cases some libraries can be missing, leading to a crash of the installation.
In this case it can be useful to edit the setup.py
file before running pip install
.