Fri, 27 Aug 2010 13:12:23 +0200 |
blanchet |
renaming + treat "TFree" better in "pattern_for_type"
|
file |
diff |
annotate
|
Fri, 27 Aug 2010 11:27:38 +0200 |
blanchet |
fix threshold computation + remove "op =" from relevant constants
|
file |
diff |
annotate
|
Thu, 26 Aug 2010 17:27:29 +0200 |
blanchet |
avoid needless "that" fact
|
file |
diff |
annotate
|
Thu, 26 Aug 2010 16:18:40 +0200 |
blanchet |
add nameless chained facts to the pool of things known to Sledgehammer
|
file |
diff |
annotate
|
Thu, 26 Aug 2010 14:58:45 +0200 |
blanchet |
if the goal contains no constants or frees, fall back on chained facts, then on local facts, etc., instead of generating a trivial ATP problem
|
file |
diff |
annotate
|
Thu, 26 Aug 2010 14:05:22 +0200 |
blanchet |
improve SPASS hack, when a clause comes from several facts
|
file |
diff |
annotate
|
Thu, 26 Aug 2010 11:51:06 +0200 |
blanchet |
lower penalty for Skolem constants
|
file |
diff |
annotate
|
Fri, 27 Aug 2010 10:56:46 +0200 |
haftmann |
formerly unnamed infix conjunction and disjunction now named HOL.conj and HOL.disj
|
file |
diff |
annotate
|
Thu, 26 Aug 2010 20:51:17 +0200 |
haftmann |
formerly unnamed infix impliciation now named HOL.implies
|
file |
diff |
annotate
|
Thu, 26 Aug 2010 10:42:06 +0200 |
blanchet |
consider "locality" when assigning weights to facts
|
file |
diff |
annotate
|
Thu, 26 Aug 2010 09:23:21 +0200 |
blanchet |
add a bonus for chained facts, since they are likely to be relevant;
|
file |
diff |
annotate
|
Thu, 26 Aug 2010 01:03:08 +0200 |
blanchet |
add a penalty for lambda-abstractions;
|
file |
diff |
annotate
|
Thu, 26 Aug 2010 00:49:04 +0200 |
blanchet |
fiddle with relevance filter
|
file |
diff |
annotate
|
Wed, 25 Aug 2010 19:41:18 +0200 |
blanchet |
reorganize options regarding to the relevance threshold and decay
|
file |
diff |
annotate
|
Wed, 25 Aug 2010 17:49:52 +0200 |
blanchet |
make relevance filter work in term of a "max_relevant" option + use Vampire SOS;
|
file |
diff |
annotate
|
Wed, 25 Aug 2010 09:42:28 +0200 |
blanchet |
simplify more code
|
file |
diff |
annotate
|
Wed, 25 Aug 2010 09:34:28 +0200 |
blanchet |
cosmetics
|
file |
diff |
annotate
|
Wed, 25 Aug 2010 09:32:43 +0200 |
blanchet |
get rid of "defs_relevant" feature;
|
file |
diff |
annotate
|
Wed, 25 Aug 2010 09:02:07 +0200 |
blanchet |
renamed "relevance_convergence" to "relevance_decay"
|
file |
diff |
annotate
|
Tue, 24 Aug 2010 22:57:22 +0200 |
blanchet |
make sure that "undo_ascii_of" is the inverse of "ascii_of", also for non-printable characters -- and avoid those in ``-style facts
|
file |
diff |
annotate
|
Tue, 24 Aug 2010 19:19:28 +0200 |
blanchet |
compute names lazily;
|
file |
diff |
annotate
|
Tue, 24 Aug 2010 18:03:43 +0200 |
blanchet |
clean handling of whether a fact is chained or not;
|
file |
diff |
annotate
|
Tue, 24 Aug 2010 16:43:52 +0200 |
blanchet |
don't backtick facts that contain schematic variables, since this doesn't work (for some reason)
|
file |
diff |
annotate
|
Tue, 24 Aug 2010 16:24:11 +0200 |
blanchet |
quote facts whose names collide with a keyword or command name (cf. "subclass" in "Jinja/J/TypeSafe.thy")
|
file |
diff |
annotate
|
Tue, 24 Aug 2010 15:07:54 +0200 |
blanchet |
cosmetics
|
file |
diff |
annotate
|
Mon, 23 Aug 2010 23:32:11 +0200 |
blanchet |
cosmetics
|
file |
diff |
annotate
|
Mon, 23 Aug 2010 23:24:24 +0200 |
blanchet |
optimize fact filter some more
|
file |
diff |
annotate
|
Mon, 23 Aug 2010 23:00:57 +0200 |
blanchet |
cache previous iteration's weights, and keep track of what's dirty and what's clean;
|
file |
diff |
annotate
|
Mon, 23 Aug 2010 21:55:52 +0200 |
blanchet |
modified relevance filter
|
file |
diff |
annotate
|
Mon, 23 Aug 2010 18:53:11 +0200 |
blanchet |
invert semantics of "relevance_convergence", to make it more intuitive
|
file |
diff |
annotate
|
Mon, 23 Aug 2010 18:39:12 +0200 |
blanchet |
if no facts were selected on first iteration, try again with a lower threshold
|
file |
diff |
annotate
|
Mon, 23 Aug 2010 18:25:49 +0200 |
blanchet |
weed out junk in relevance filter
|
file |
diff |
annotate
|
Mon, 23 Aug 2010 18:04:31 +0200 |
blanchet |
no need for "eq" facts -- good old "=" is good enough for us
|
file |
diff |
annotate
|
Mon, 23 Aug 2010 17:49:18 +0200 |
blanchet |
destroy elim rules before checking for finite exhaustive facts
|
file |
diff |
annotate
|
Mon, 23 Aug 2010 14:54:17 +0200 |
blanchet |
perform eta-expansion of quantifier bodies in Sledgehammer translation when needed + transform elim rules later;
|
file |
diff |
annotate
|
Sun, 22 Aug 2010 16:56:05 +0200 |
blanchet |
don't penalize abstractions in relevance filter + support nameless `foo`-style facts
|
file |
diff |
annotate
|
Sun, 22 Aug 2010 08:26:09 +0200 |
blanchet |
more work on finite axiom detection
|
file |
diff |
annotate
|
Fri, 20 Aug 2010 17:52:26 +0200 |
blanchet |
make sure minimizer facts go through "transform_elim_theorems"
|
file |
diff |
annotate
|
Fri, 20 Aug 2010 16:44:48 +0200 |
blanchet |
unbreak "only" option of Sledgehammer
|
file |
diff |
annotate
|
Fri, 20 Aug 2010 16:22:51 +0200 |
blanchet |
improve "x = A | x = B | x = C"-style axiom detection
|
file |
diff |
annotate
|
Fri, 20 Aug 2010 14:15:29 +0200 |
blanchet |
improve "x = A | x = B | x = C"-style fact discovery
|
file |
diff |
annotate
|
Fri, 20 Aug 2010 14:09:02 +0200 |
blanchet |
transform elim theorems before filtering "bool" and "prop" variables out;
|
file |
diff |
annotate
|
Fri, 20 Aug 2010 13:39:47 +0200 |
blanchet |
more fiddling with Sledgehammer's "add:" option
|
file |
diff |
annotate
|
Thu, 19 Aug 2010 19:34:11 +0200 |
blanchet |
generalize the "too general equality" code to handle facts like "x ~= A ==> x = B"
|
file |
diff |
annotate
|
Thu, 19 Aug 2010 18:16:47 +0200 |
blanchet |
encode "fequal" reasoning rules in Metis problem, just as is done for Sledgehammer -- otherwise any proof that relies on "fequal" found by Sledgehammer can't be reconstructed
|
file |
diff |
annotate
|
Wed, 18 Aug 2010 20:17:03 +0200 |
blanchet |
make sure that "add:" doesn't influence the relevance filter too much
|
file |
diff |
annotate
|
Wed, 18 Aug 2010 19:57:12 +0200 |
blanchet |
tuning of relevance filter
|
file |
diff |
annotate
|
Wed, 18 Aug 2010 18:01:54 +0200 |
blanchet |
minor cleanup
|
file |
diff |
annotate
|
Wed, 18 Aug 2010 17:53:12 +0200 |
blanchet |
bound variables can be just as evil as schematic variables and lead to unsound proofs (e.g. "all_bool_eq")
|
file |
diff |
annotate
|
Wed, 18 Aug 2010 16:39:05 +0200 |
blanchet |
fix the relevance filter so that it ignores If, Ex1, Ball, Bex
|
file |
diff |
annotate
|
Thu, 12 Aug 2010 17:56:44 +0200 |
haftmann |
dropped dead code
|
file |
diff |
annotate
|
Mon, 09 Aug 2010 14:08:30 +0200 |
blanchet |
prevent ATP thread for staying around for 1 minute if an exception occurred earlier;
|
file |
diff |
annotate
|
Mon, 09 Aug 2010 14:00:32 +0200 |
blanchet |
adapt "too_general_equality" blacklisting to the new FOF context, where quantifiers are sometimes present
|
file |
diff |
annotate
|
Mon, 09 Aug 2010 10:39:53 +0200 |
blanchet |
remove debugging output
|
file |
diff |
annotate
|
Thu, 29 Jul 2010 22:57:36 +0200 |
blanchet |
handle division by zero gracefully (used to raise Unordered later on)
|
file |
diff |
annotate
|
Thu, 29 Jul 2010 19:58:43 +0200 |
blanchet |
avoid "_def_raw" theorems
|
file |
diff |
annotate
|
Thu, 29 Jul 2010 17:45:22 +0200 |
blanchet |
work around atomization failures
|
file |
diff |
annotate
|
Thu, 29 Jul 2010 16:54:46 +0200 |
blanchet |
fiddle with the fudge factors, to get similar results as before
|
file |
diff |
annotate
|
Thu, 29 Jul 2010 14:53:55 +0200 |
blanchet |
avoid "clause" and "cnf" terminology where it no longer makes sense
|
file |
diff |
annotate
|
Tue, 27 Jul 2010 19:17:15 +0200 |
blanchet |
standardize "Author" tags
|
file |
diff |
annotate
|