src/HOL/Tools/Sledgehammer/sledgehammer_run.ML
Thu, 31 Jan 2013 17:54:05 +0100 blanchet tuned data structure
Thu, 31 Jan 2013 17:54:05 +0100 blanchet thread through fact triple component from which used facts come, for accurate index output
Thu, 31 Jan 2013 17:54:05 +0100 blanchet more precise output of selected facts
Thu, 31 Jan 2013 17:54:05 +0100 blanchet thread fact triple (MeSh, MePo, MaSh) to allow different filters in different slices
Thu, 31 Jan 2013 17:54:05 +0100 blanchet simplified SMT solver code in Sledgehammer
Thu, 31 Jan 2013 17:54:05 +0100 blanchet eliminated needless speed optimization -- and simplified code quite a bit
Thu, 31 Jan 2013 17:54:05 +0100 blanchet distinguish raw and non-raw facts, using raw for 10 000s of facts and non-raw after selection of some hundreds
Thu, 31 Jan 2013 17:54:05 +0100 blanchet report (MeSh, MePo, MaSh) triple, to be able to use different filters in different slices
Tue, 15 Jan 2013 20:51:30 +0100 blanchet tuned whitespace
Mon, 14 Jan 2013 10:32:33 +0100 blanchet run Sledgehammer provers in parallel in "try"
Sun, 13 Jan 2013 21:42:38 +0100 blanchet tuned message
Sat, 05 Jan 2013 22:31:31 +0100 blanchet increased hard timeout -- minimization can take time
Wed, 02 Jan 2013 10:54:36 +0100 blanchet use "Lazy" to simplify control flow a bit and guarantee single evaluation (at most)
Wed, 02 Jan 2013 10:41:53 +0100 blanchet tuning
Sat, 15 Dec 2012 19:57:12 +0100 blanchet thread no timeout properly
Sun, 25 Nov 2012 19:49:24 +0100 wenzelm Isabelle-specific implementation of quasi-abstract markup elements -- back to module arrangement before d83797ef0d2d;
Thu, 18 Oct 2012 13:19:44 +0200 blanchet refactor code
Fri, 14 Sep 2012 18:12:41 +0200 wenzelm clarified markup names;
Tue, 14 Aug 2012 13:20:59 +0200 blanchet be less aggressive at kicking out chained facts
Fri, 20 Jul 2012 22:19:46 +0200 blanchet faster maximal node computation
Fri, 20 Jul 2012 22:19:46 +0200 blanchet learn from SMT proofs when they can be minimized by Metis
Fri, 20 Jul 2012 22:19:46 +0200 blanchet name tuning
Fri, 20 Jul 2012 22:19:46 +0200 blanchet fixed various issues with MaSh's file handling + tune output + generate local facts again + handle nameless facts gracefully
Fri, 20 Jul 2012 22:19:46 +0200 blanchet learn on explicit "min" command but do the learning in a thread, since it may take a couple of seconds
Fri, 20 Jul 2012 22:19:45 +0200 blanchet renamed ML structures
Wed, 18 Jul 2012 08:44:04 +0200 blanchet learn from minimized ATP proofs
Wed, 18 Jul 2012 08:44:04 +0200 blanchet use async manager to manage MaSh learners to make sure they get killed cleanly
Wed, 18 Jul 2012 08:44:03 +0200 blanchet centrally construct expensive data structures
Wed, 18 Jul 2012 08:44:03 +0200 blanchet renamed Sledgehammer options
Wed, 18 Jul 2012 08:44:03 +0200 blanchet more code rationalization in relevance filter
Wed, 18 Jul 2012 08:44:03 +0200 blanchet systematize lazy names in relevance filter
Wed, 18 Jul 2012 08:44:03 +0200 blanchet rationalize relevance filter, slowing moving code from Iter to MaSh
Wed, 11 Jul 2012 21:43:19 +0200 blanchet further ML structure split to permit finer-grained loading/reordering (problem to solve: MaSh needs most of Sledgehammer)
Thu, 10 May 2012 16:56:23 +0200 blanchet distinguish between instantiated and uninstantiated inductions -- the latter are OK for first-order provers
Wed, 18 Apr 2012 10:53:28 +0200 blanchet get rid of minor optimization that caused strange problems and was hard to debug (and apparently saved less than 100 ms on a 30 s run)
Tue, 27 Mar 2012 16:59:13 +0300 blanchet be less forceful about ":lt" to make infinite loops less likely (could still fail with mutually recursive tail rec functions)
Mon, 12 Mar 2012 23:16:54 +0100 wenzelm Par_List.map is already smart;
Sat, 04 Feb 2012 12:08:18 +0100 blanchet made option available to users (mostly for experiments)
Thu, 26 Jan 2012 20:49:54 +0100 blanchet separate orthogonal components
Mon, 23 Jan 2012 17:40:32 +0100 blanchet renamed two files to make room for a new file
Thu, 19 Jan 2012 21:37:12 +0100 blanchet renamed "sound" option to "strict"
Thu, 01 Dec 2011 13:34:14 +0100 blanchet added "minimize" option for more control over automatic minimization
Thu, 01 Dec 2011 13:34:13 +0100 blanchet renamed "slicing" to "slice"
Mon, 28 Nov 2011 22:05:32 +0100 wenzelm separate module for concrete Isabelle markup;
Fri, 18 Nov 2011 11:47:12 +0100 blanchet don't propagate user-set "type_enc" or "lam_trans" to Metis calls
Wed, 16 Nov 2011 16:35:19 +0100 blanchet thread in additional options to minimizer
Sun, 06 Nov 2011 22:18:54 +0100 blanchet try "smt" as a fallback for ATPs if "metis" fails/times out
Sun, 06 Nov 2011 13:57:17 +0100 blanchet use "Time.time" rather than milliseconds internally
Sun, 06 Nov 2011 13:46:26 +0100 blanchet tune: no need for option
Mon, 12 Sep 2011 10:49:37 +0200 blanchet consistent option naming
Wed, 31 Aug 2011 11:52:03 +0200 blanchet more tuning
Tue, 30 Aug 2011 14:12:55 +0200 nik improved handling of induction rules in Sledgehammer
Tue, 30 Aug 2011 14:12:55 +0200 nik added generation of induction rules
Tue, 28 Jun 2011 21:06:59 +0200 blanchet reenabled accidentally-disabled automatic minimization
Fri, 10 Jun 2011 12:01:15 +0200 blanchet compute the set of base facts only once (instead of three times in parallel) -- this saves about .5 s of CPU time, albeit much less clock wall time
Thu, 09 Jun 2011 00:16:28 +0200 blanchet tuning
Wed, 08 Jun 2011 16:20:18 +0200 blanchet don't launch the automatic minimizer for zero facts
Wed, 08 Jun 2011 16:20:18 +0200 blanchet renamed option to avoid talking about seconds, since this is now the default Isabelle unit
Wed, 08 Jun 2011 08:47:43 +0200 blanchet make sure that the message tail (timing + TPTP important message) is preserved upon automatic minimization
Tue, 07 Jun 2011 11:04:17 +0200 blanchet use the proper prover name, e.g. metis_full_types, not metis (full_types), for minimizing
less more (0) -100 -60 tip