Developer SetupΒΆ

The prerequisites are:

  • Python 3, either

    • system-wide installation with virtualenv, or
    • installed with Conda.
  • Git LFS must be installed for obtaining the example data files.

For virtualenv, first create a virtual environment and activate it.

$ virtualenv -p venv
$ source venv/bin/activate

For a Conda-based setup create a new environment and activate it.

$ conda create -y -n scelvis 'python>=3.6'
$ conda activate scelvis

Next, clone the repository and install the software as editable (-e). Also install the development requirements to get helpers such as black. (Note that you must have Git LFS installed to actually obtain the data files).

$ git clone
$ cd scelvis
$ pip install -e .
$ pip install -r requirements/develop.txt

Afterwards, you can run the visualization web server as follows:

$ scelvis run --data-source path/to/data/dir

To explore the datasets provided in the git repository, use git lfs fetch to download