h20zhang/frasmt

FraSMT: SMT-based decomposer for fractional hypertree decompositions

FraSMT (SMT-based decomposer for fractional hypertree decompositions)

Download:

git clone --recurse-submodules  [email protected]:daajoe/frasmt.git

Hypergraphs

Fischl (hyperbench collection)

See: https://github.com/daajoe/hypergraphs or http://hyperbench.dbai.tuwien.ac.at/

External Requirements (conda)

Get Anaconda

https://www.anaconda.com/download/#macos

Remark

Conda is a little memory hungry. So you want to give your system at least 1G memory. If installation quits just quits, run with --verbose option. In case you run into "CondaMemoryError does not take keyword arguments", then you really want to increase available memory to conda.

Setup Environment

conda update -n base conda
conda env create -f environment.yml  --name fhtd #(meanwhile get a yourself coffee; or two)

Active Environment

source activate fhtd

SMT solver (non-linux)

Unfortunately, anaconda packages for z3 are quite outdated (we need the optimization API from z3). I put together an experimental anaconda package (https://anaconda.org/daajoe/z3_experimental), but do not maintain this one at the moment. Should work, but only for linux. Otherwise see below.

optimathsat

http://optimathsat.disi.unitn.it/pages/download-js.html

Z3

https://github.com/Z3Prover/z3

with python support

python scripts/mk_make.py --prefix=/home/USERNAME/miniconda2 --python --pypkgdir=/home/USERNAME/miniconda2/lib/python-2.7/site-packages cd build make -j4 make install

External Requirements (not using conda)

We highly recommend to use conda!!

pysmt (optional)

pysmt optimization open PR: pysmt/pysmt#439

clingo

• conda: https://conda.io/docs/user-guide/tasks/manage-pkgs.html
• https://anaconda.org/potassco/clingo

cplex

https://www.ibm.com/support/knowledgecenter/es/SSSA5P_12.6.1/ilog.odms.studio.help/Optimization_Studio/topics/COS_installing.html https://www.ibm.com/support/knowledgecenter/SSSA5P_12.7.0/ilog.odms.cplex.help/CPLEX/GettingStarted/topics/set_up/Python_setup.html

• Install Anaconda cplex from https://anaconda.org/IBMDecisionOptimization/cplex
• Download cplex (IBM Academic Initative)
  • Install it into your home (/home/myhome/cplex_1210)
  • Install python into local site packages somewhere (not into anaconda directory; https://or.stackexchange.com/questions/3114/python-does-not-identify-the-academic-version-of-cplex)
  • cp ~/cplex_1210/cplex/python/3.7/x86-64_linux/cplex/_internal/py37_cplex12100.so ~/anaconda3/envs/rb/lib/python3.7/site-packages/cplex/_internal/py37_cplex12100.so

Test it

bin/fhtd -f tests/graphs/easy/c4.hg