Installation

PyPI Installation

TAT-C uses pip and the PyPI software repository to install the main package along with underlying dependencies and libraries.

TAT-C can be installed from PyPI via the terminal/shell command pip install tatc.

Then, TAT-C is available for use in any Python script by importing import tatc.

Conda Installation

TAT-C uses conda and the conda-forge channel for distribution because some of the underlying libraries are platform dependent.

The simplest way to install TAT-C is via the terminal/shell command conda install conda-forge::tatc.

Then, TAT-C is available for use in any Python script by importing import tatc.

Source Installation

Alternatively, TAT-C can be installed from a local editable source, (e.g., for modifying TAT-C functionality). This method is required to access run the web-based application.

Clone the project repository using git clone https://github.com/code-lab-org/tatc.git.

Install the tatc library in “editable” mode pip install -e .

Note: the following optional dependencies are available with bracket notation:

pip install -e ".[dev]": for development functions (unit testing, coverage, and linting)
pip install -e ".[docs]": for generating documentation in docs/
pip install -e ".[examples]": for running optional examples in docs/examples
pip install -e ".[app]": for running the web application

Multiple optional dependencies can be installed with a comma-separated list (e.g., pip install -e ".[dev,examples]")

Faster Package Resolution

Dependency resolution in conda can be very slow. For faster installation, consider installing the mamba package conda install mamba -c conda-forge and replace conda with mamba in the installation instructions above.

Unit Tests

Run unit tests from the project root directory with python -m unittest or coverage run -m unittest.

Documentation

Build the documentation from the docs/ directory with make html.