Replication package for Efficient Slicing of Feature Models via Projected d-DNNF Compilation (ASE'24)

Steps to reproduce

Cloning

The repository includes pd4 as submodule. To clone please use:

git clone --recurse-submodules [email protected]:SoftVarE-Group/proj-ddnnf-eval.git

The version of pd4 used originally in this evaluated is tagged with ase24 in https://github.com/SoftVarE-Group/pd4.

Build Solvers

• Original d4, slicing, and arjun are provided as binaries
• Please follow the instructions in the pd4 submodule to install required dependencies for pd4 and compile it

Get the feature models

Due to their size, the feature models are provided at Zenodo: https://zenodo.org/records/13752110

The expected paths are for running the eval are input/fm-gen/ and input/MC2022_track3-pmc_private/

Setup

pip3 install -r requirements.txt

Run evaluation

Industrial models: Left out due to confidentiality

Literature Models: python3 run.py run_configurations/literaturemodels.yaml

Competition models: python3 run.py run_configurations/competitionnstances.yaml