FFNSL experiments to support the paper "FFNSL: Feed-Forward Neural-Symbolic Learner". This code has been tested on MacOSX Big Sur 11.6.1.

• python >= 3.7
• ILASP including clingo.
• FastLAS
• Graphviz

From the project root:

1. pip install -r requirements.txt
2. python setup.py

Data and pre trained models are available to download here. Extract the zip folder and copy the files into the directory tree.

Refer to the run_experiments.md file in each example directory. Results presented in the FFNSL paper are available in the paper_results folder of each example.