Tarski - The Planning Problem Definition Module

What is Tarski

Tarski is a parser and preprocessor to create, manipulate and transform AI planning problems.

Installation

You can install the latest official Tarski release with

pip install tarski

If instead you want to use the latest code available on the Github repository, use

pip install -U git+https://github.com/aig-upf/tarski.git

Development

If developing Tarski, we recommend cloning from the Github repository and doing a development installation (pip's -e flag) on a virtual environment:

git clone https://github.com/aig-upf/tarski
cd tarski
pip install -e .

This will install the project in "editable mode", i.e., by using the actual files from the project directory tree, as opposed to a copy installed in some system or virtualenv directory. Any modification to the files will thus be immediately reflected in the installed library.

Testing

All of Tarski's tests live under the tests directory (shocking!). To run them, you just need to run pytest (pip install pytest) on the root directory. You can also run tox (pip install tox) to have some additional checks (e.g., style checks) run.

Software Requirements

Tarski requires Python >= 3.5. The above installation instructions will install under the hood a number of dependencies, among which numpy and scipy.

Supported Languages

At the moment, Tarski is able to parse the following modeling languages:

• PDDL
• Functional STRIPS

with the following limitations:

• The PDDL either keyword for defining compound types is not supported.
• Parsing of derived predicates is not supported yet.

Modules

• Description Logic (tarski.dl)
• PDDL / FSTRIPS parsing and writing (tarski.io)