Closed vhavlena closed 2 months ago
The results seem ok:
# of formulae: 103318
--------------------------------------------------
tool ✅ ❌ time avg med std sat unsat unknown TO MO+ERR other
-------------------------- ------ ---- --------- ----- ----- ----- ----- ------- --------- ---- -------- -------
z3-noodler-ad39615-d95fe13 102228 1090 12781.90 0.13 0.02 1.69 62798 39430 625 406 59 0
z3-noodler-69644b5-d95fe13 102231 1087 11660.24 0.11 0.01 1.57 62805 39426 625 411 51 0
cvc5-1.1.2 99699 3619 76478.24 0.77 0.00 6.77 60798 38901 2 3608 9 0
z3-4.13.0 97687 5631 132036.99 1.35 0.01 8.11 58874 38813 210 4960 461 0
--------------------------------------------------
I think we can merge this. I would run the experiments again after merging, because #152 is not merged here.
I am reverting the last commit, as it unexpectedly caused some performance degradation (not sure why and there is no manpower to find it out).
This PR fixes bug in preprocessing. The preprocessing rule
skip_len_sat
requires to be the initial formula length satisfiable. Therefore, I moved the initial length check before the preprocessing itself to ensure that the formula meets the preprocessing requirements.