Skip to content

Installing pySNOW

Installing pySNOW is (should be) quite easy.

The first step (suggested) is to create a python environment for installation to avoid confilcts with previously installed libraries or different versions of required modules:

python3 -m venv snowenv
source snowenv/bin/activate

python3 -m venv snowenv
source snowenv/bin/activate

python -m venv snowenv
.\snowenv\Scripts\activate

Note that pySNOW b default requires a version of python >= 3.9, however users can optionally try to change this in the pyproject.toml file, this however has not been tried and might result in unwanted behaviour. Going below python 3 is sure to cause issues due to syntax differences.

After creating and activating the virtual environment you can clone the repository (or download the code as a zip and uncompressing it in any deisred location):

git clone https://github.com/nanoMLMS/pySNOW.git pysnow && cd pysnow

From here you should be able to easily install the package by running: 

pip install .

Automated tests can be run using pytest with the following command: 

pytest .
from the pysnow folder.

Requirments for pYSNOW are NumPy and SciPy, which are automatically installed with pySNOW, additional optional dependencies are:

  • tqdm for progress bars to check progress of long computations
  • ase for advanced input/output and exploitation of functions implemented there (note that ase is based on classes encapsulating structures rather than simple arrays)