Re: crm114 on 64 bit SuSE
Thomas Spahni <[email protected]>
| Newsgroups | gmane.mail.spam.crm114 |
|---|---|
| Message-ID | <[email protected]> |
On Mon, 11 Aug 2008, Ger Hobbelt wrote: > Thanks for testing and reporting back; much appreciated! You're welcome. > > Further responses mingled with original email quotes: > > On Sun, Aug 10, 2008 at 2:07 PM, Thomas Spahni <[email protected]> wrote: > [..] > > mail) but the build works well in production. There is just one glitch I > > observed: learning just one message advances the count of messages learned > > by TWO. I still could not find the reason for this. Any pointers would be > > welcome. > > Which classifier did you use? (Each keeps track of learn counts > individually, hence the question) My mailfilter.cf selects: :clf: /osb unique microgroom/ > > > I tried to install Ger's crm114-20080502-BlameJason-Ger-2423.src.tar as > > well. Compile went well but learning a message with mailreaver results in: > > Whoops. Hrmmm... yeah; I didn't test mail*.crm scripts for a long time > as I don't use them myself (I use crm114 for different purposes). > I see where it could produce this kind of barfball, but I am very > short on time now, so fixing this will have to wait unfortunately. > :-(( > > Mea maxima culpa. Don't bother. It's open source and when I'm too stupid to fix it myself I don't expect the developers to tackle the problem immediately. I will give it a try myself and report. > > [...] > > Additionally Ger's version produces a significant list of differences in > > megatest (output attached). > > Part of the diffs are the 'extra titles' (which show up twice instead > of once) so that I can quickly see which classifiers produced what > diff bits; important when you are looking at slews of numbers. > > Let's have a look at your megatest output. > (By the way, did you try 'make check' as well? That is ONLY available > with the GerH builds and runs a growing series of tests which span > more than megatest does. I'm interested to hear what that would > produce on your system...) Here it goes: Making check in docs make[1]: Entering directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/docs' make[1]: Nothing to be done for `check'. make[1]: Leaving directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/docs' Making check in examples make[1]: Entering directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/examples' make[1]: Nothing to be done for `check'. make[1]: Leaving directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/examples' Making check in mailfilter make[1]: Entering directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/mailfilter' make[1]: Nothing to be done for `check'. make[1]: Leaving directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/mailfilter' Making check in man make[1]: Entering directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/man' make[1]: Nothing to be done for `check'. make[1]: Leaving directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/man' Making check in src make[1]: Entering directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/src' make check-TESTS make[2]: Entering directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/src' ================== All 0 tests passed ================== make[2]: Leaving directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/src' make[1]: Leaving directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/src' Making check in tests make[1]: Entering directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/tests' Makefile:1908: warning: overriding commands for target `check-am' Makefile:847: warning: ignoring old commands for target `check-am' make[2]: Entering directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/tests' Makefile:1908: warning: overriding commands for target `check-am' Makefile:847: warning: ignoring old commands for target `check-am' make[2]: Nothing to be done for `all'. make[2]: Leaving directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/tests' make[2]: Entering directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/tests' Makefile:1908: warning: overriding commands for target `check-am' Makefile:847: warning: ignoring old commands for target `check-am' ================== All 0 tests passed ================== /bin/mkdir -p ./ref Creating diff archives (cd ../src && make all); make[3]: Entering directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/src' make all-am make[4]: Entering directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/src' make[4]: Leaving directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/src' make[3]: Leaving directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/src' test: CRM114 '-{}' -v targetname = ./crm114_versiontest.filter.sh.temp OK test: CRM114 'bracktest.crm' OK test: CRM114 'alius_w_comment.crm' OK test: CRM114 'escapetest.crm' OK test: CRM114 'fataltraptest.crm' targetname = ./crm114_error_report.filter.sh.temp OK test: CRM114 'fataltrapprogramerr.crm' targetname = ./crm114_error_report.filter.sh.temp OK test: CRM114 'inserttest_a.crm' OK test: CRM114 'matchtest.crm' targetname = ./crm114_error_report.filter.sh.temp OK test: CRM114 'backwardstest.crm' OK test: CRM114 'backwardstest.crm' OK test: CRM114 'overalterisolatedtest.crm' OK test: CRM114 'rewritetest.crm' OK test: CRM114 'skudtest.crm' OK test: CRM114 'statustest.crm' targetname = ./crm114_statustest.filter.sh.temp OK test: CRM114 'exectest.crm' targetname = ./crm114_statustest.filter.sh.temp OK test: CRM114 'unionintersecttest.crm' OK test: CRM114 'beeptest.crm' OK test: CRM114 'defaulttest.crm' OK test: CRM114 'defaulttest.crm' --blah=command override OK test: CRM114 'windowtest.crm' OK test: CRM114 'windowtest_fromvar.crm' OK test: CRM114 'approxtest.crm' OK test: CRM114 'mathalgtest.crm' OK test: CRM114 'mathrpntest.crm' -q 1 OK test: CRM114 'eval_infiniteloop.crm' targetname = ./crm114_error_report.filter.sh.temp OK test: CRM114 'randomiotest.crm' OK test: CRM114 'paolo_overvars.crm' OK test: CRM114 'paolo_ov2.crm' OK test: CRM114 'paolo_ov3.crm' OK test: CRM114 'paolo_ov4.crm' OK test: CRM114 'paolo_ov5.crm' OK test: CRM114 'print_binary2decimal_int32.crm' OK test: CRM114 'match_re_fringe1.crm' OK !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! This runs a moderately interesting set of base tests to exercise much of CRM114 under TRE. This takes several seconds to run on a 3 GHz AMD/XP machine. Please be patient; you (hopefully) won't see anything till the full set of tests complete. If you didn't use TRE, all bets are off. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! test: CRM114 'match_isolate_test.crm' -e targetname = ./crm114_isolate_test.filter.sh.temp OK test: CRM114 'match_isolate_reclaim.crm' -e targetname = ./crm114_isolate_test.filter.sh.temp OK test: CRM114 'call_return_test.crm' OK test: CRM114 'translate_tr.crm' OK test: CRM114 'zz_translate_test.crm' OK test: CRM114 'quine.crm' OK test: CRM114 '-{window; isolate (:s:); syscall () (:s:) /echo one two three/; output /:*:s:/}' OK test: CRM114 '-{window; isolate (:s:); syscall () () (:s:) /exit 123/; output / Status: :*:s: \n/}' OK test: CRM114 '-{window; isolate (:s:); syscall () () (:s:) /jibberjabber 2>&1 /; output / Status: :*:s: \n/}' OK test: CRM114 'indirecttest.crm' OK ******************************************* * Default (SBPH Markovian) classifier ******************************************* rm -f randtst.txt rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:) {classify ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: OK test: CRM114 '-{ isolate (:s:) {classify ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: OK ******************************************* * OSB Markovian classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn <osb> (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn <osb> (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify <osb> ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' OK test: CRM114 '-{ isolate (:s:); {classify <osb> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' OK ******************************************* * OSB Markov Unique classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn <osb unique > (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn <osb unique > (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify <osb unique> ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' OK test: CRM114 '-{ isolate (:s:); {classify <osb unique> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' OK ******************************************* * OSB Markov Chisquared Unique classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn <osb unique chi2> (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn <osb unique chi2> (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify <osb unique chi2 > ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' OK test: CRM114 '-{ isolate (:s:); {classify <osb unique chi2> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' OK ******************************************* * OSBF Local Confidence (Fidelis) classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn < osbf > (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn < osbf > (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify <osbf> ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' OK test: CRM114 '-{ isolate (:s:); {classify <osbf> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' OK ******************************************* * OSB Winnow classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn <winnow> (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn <winnow refute> (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn <winnow> (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn <winnow refute> (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify <winnow> ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } } ' OK test: CRM114 '-{ isolate (:s:); {classify <winnow> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }} ' OK Now verify that winnow learns affect only the named file (i_test.css) test: CRM114 '-{learn <winnow> (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify <winnow> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }} ' OK and now refute-learn into q_test.css test: CRM114 '-{learn <winnow refute > (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify <winnow> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }} ' OK ******************************************* * Unigram Bayesian classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn <unigram> (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn <unigram> (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify <unigram> ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' OK test: CRM114 '-{ isolate (:s:); {classify <unigram> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' OK ******************************************* * unigram Winnow classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn <winnow unigram > (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn <winnow unigram refute> (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn <winnow unigram> (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn <winnow unigram refute> (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify <winnow unigram> ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } } ' OK test: CRM114 '-{ isolate (:s:); {classify <winnow unigram> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }} ' OK ******************************************* * OSB Hyperspace classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn <hyperspace unique> (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn <hyperspace unique> (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify <hyperspace> ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' OK test: CRM114 '-{ isolate (:s:); {classify <hyperspace> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' OK ******************************************* * OSB three-letter Hyperspace classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn <hyperspace unique> (q_test.css) /.../}' OK test: CRM114 '-{learn <hyperspace unique> (i_test.css) /.../}' OK test: CRM114 '-{ isolate (:s:); {classify <hyperspace> ( i_test.css | q_test.css ) (:s:) /.../ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' OK test: CRM114 '-{ isolate (:s:); {classify <hyperspace> ( i_test.css | q_test.css ) (:s:) /.../ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' OK ******************************************* * Unigram Hyperspace classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn < hyperspace unique unigram> (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn < hyperspace unique unigram> (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify < hyperspace unigram> ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' OK test: CRM114 '-{ isolate (:s:); {classify <hyperspace unigram> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' OK ******************************************* * String Hyperspace classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn < hyperspace string> (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn < hyperspace string> (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify < hyperspace string> ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' OK test: CRM114 '-{ isolate (:s:); {classify <hyperspace string> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' OK ******************************************* * String Unigram Hyperspace classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn < hyperspace string unigram> (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn < hyperspace string unigram> (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify < hyperspace string unigram> ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' OK test: CRM114 '-{ isolate (:s:); {classify <hyperspace string unigram> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' OK ******************************************* * Vector 3-word-bag Hyperspace classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn < hyperspace > (q_test.css) /[[:graph:]]+/ /vector: 3 1 1 1 1 1 / }' OK test: CRM114 '-{learn < hyperspace > (i_test.css) /[[:graph:]]+/ /vector: 3 1 1 1 1 1/}' OK test: CRM114 '-{ isolate (:s:); {classify < hyperspace > ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ /vector: 3 1 1 1 1 1 /; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' OK test: CRM114 '-{ isolate (:s:); {classify <hyperspace > ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ /vector: 3 1 1 1 1 1 /; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' OK ******************************************* * Bit-Entropy classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn < entropy unique crosslink> (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn < entropy unique crosslink> (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify < entropy unique crosslink> ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v entropy=50 -v o_entropy=370 -v jumps=4 -v o_jumps=7 -v prob=2.0 -v pR=3.0 OK test: CRM114 '-{ isolate (:s:); {classify <entropy unique crosslink> ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v file_features=300 -v entropy=50 -v o_entropy=370 -v jumps=40 -v o_jumps=7 -v prob=5.0 -v pR=5.0 ---- diff report: ---- --- ./ref/Bit_Entropy_test1.step4.refoutput 2008-05-01 08:37:58.000000000 +0200 +++ ./testscript.sh.tempchk 2008-08-12 10:26:46.000000000 +0200 @@ -7,11 +7,11 @@ STDERR TEST OUTPUT: STDOUT TEST OUTPUT: - type Q -CLASSIFY fails; success probability: 0.0000 pR: -6.5308 -Best match to file #1 (q_test.css) prob: 1.0000 pR: 6.5308 -Total features in input file: 552 -#0 (i_test.css): features: 425488 (100%), entropy: 459.942001, jumps: 26, prob: 2.95e-07, pR: -6.53 -#1 (q_test.css): features: 552504 (100%), entropy: 302.166313, jumps: 10, prob: 1.00e+00, pR: 6.53 + type I +CLASSIFY succeeds; success probability: 1.0000 pR: 4.9105 +Best match to file #0 (i_test.css) prob: 1.0000 pR: 4.9105 +Total features in input file: 408 +#0 (i_test.css): features: 425488 (5%), entropy: 138.295563, jumps: 4, prob: 1.00e-00, pR: 4.91 +#1 (q_test.css): features: 552504 (5%), entropy: 256.926733, jumps: 16, prob: 1.23e-05, pR: -4.91 ---------------------- FAIL(OVERRIDEN:OK) ******************************************* * Bit-Entropy Toroid classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn < entropy > (q_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{learn < entropy > (i_test.css) /[[:graph:]]+/}' OK test: CRM114 '-{ isolate (:s:); {classify < entropy > ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v entropy=100 -v jumps=10 -v prob=1.0 -v pR=4.0 OK test: CRM114 '-{ isolate (:s:); {classify < entropy > ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v entropy=25 -v jumps=2 -v prob=1.0e-5 -v pR=1.0 OK ******************************************* * Fast Substring Compression Match Classifier ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{learn < fscm > (q_test.css) /[[:graph:]]+/}' -s 200000 OK test: CRM114 '-{learn < fscm > (i_test.css) /[[:graph:]]+/}' -s 200000 OK test: CRM114 '-{ isolate (:s:); {classify < fscm > ( i_test.css | q_test.css ) (:s:)/[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' OK test: CRM114 '-{ isolate (:s:); {classify < fscm > ( i_test.css | q_test.css ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' OK ******************************************* * Support Vector Machine (SVM) unigram classifier ******************************************* rm -f i_vs_q_test.css rm -f i_test.css rm -f q_test.css test: CRM114 '-{ match <fromend> (:one_paragraph:) /[[:graph:]]+.*?\n\n/; learn [:one_paragraph:] < svm unigram unique > (i_test.css) /[[:graph:]]+/; liaf}' OK test: CRM114 '-{ match <fromend> (:one_paragraph:) /[[:graph:]]+.*?\n\n/; learn [:one_paragraph:] < svm unigram unique > (q_test.css) /[[:graph:]]+/; liaf }' OK build the actual hyperplanes test: CRM114 '-{window; learn ( i_test.css | q_test.css | i_vs_q_test.css ) < svm unigram unique > /[[:graph:]]+/ /0 0 100 1e-3 1 0.5 1 1/ }' OK test: CRM114 '-{ isolate (:s:); {classify < svm unigram unique > ( i_test.css | q_test.css | i_vs_q_test.css ) (:s:) /[[:graph:]]+/ /0 0 100 1e-3 1 0.5 1 1/ [:_dw:] ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v prob=0.1 -v pR=0.2 OK test: CRM114 '-{ isolate (:s:); {classify < svm unigram unique > ( i_test.css | q_test.css | i_vs_q_test.css ) (:s:) /[[:graph:]]+/ /0 0 100 1e-3 1 0.5 1 1/ [:_dw:] ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v prob=1e-4 -v pR=0.03 -vo_pR=22.74 OK ******************************************* * Support Vector Machine (SVM) classifier ******************************************* rm -f i_vs_q_test.css rm -f i_test.css rm -f q_test.css test: CRM114 '-{ match <fromend> (:one_paragraph:) /[[:graph:]]+.*?\n\n/; learn [:one_paragraph:] < svm unique > (i_test.css) /[[:graph:]]+/; liaf}' OK test: CRM114 '-{ match <fromend> (:one_paragraph:) /[[:graph:]]+.*?\n\n/; learn [:one_paragraph:] < svm unique > (q_test.css) /[[:graph:]]+/; liaf }' OK build the actual hyperplanes test: CRM114 '-{window; learn ( i_test.css | q_test.css | i_vs_q_test.css ) < svm unique > /[[:graph:]]+/ /0 0 100 1e-3 1 0.5 1 1/ }' OK test: CRM114 '-{ isolate (:s:); {classify < svm unique > ( i_test.css | q_test.css | i_vs_q_test.css ) (:s:) /[[:graph:]]+/ /0 0 100 1e-3 1 0.5 1 1/ [:_dw:] ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v prob=4e-3 -v o_prob=0.1275 -v pR=0.04 -v o_pR=8.36 OK test: CRM114 '-{ isolate (:s:); {classify < svm unique > ( i_test.css | q_test.css | i_vs_q_test.css ) (:s:) /[[:graph:]]+/ /0 0 100 1e-3 1 0.5 1 1/ [:_dw:] ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v prob=2e-4 -v pR=0.04 -v o_pR=14.775 OK ******************************************* * String Kernel SVM (SKS) classifier ******************************************* rm -f i_vs_q_test.css rm -f i_test.css rm -f q_test.css test: CRM114 '-{ match <fromend> (:one_paragraph:) /[[:graph:]]+.*?\n\n/; learn [:one_paragraph:] < sks > (i_test.css) /[[:graph:]]+/; liaf}' OK test: CRM114 '-{ match <fromend> (:one_paragraph:) /[[:graph:]]+.*?\n\n/; learn [:one_paragraph:] < sks > (q_test.css) /[[:graph:]]+/; liaf }' OK build the actual hyperplanes test: CRM114 '-{window; learn ( i_test.css | q_test.css | i_vs_q_test.css ) < sks > /[[:graph:]]+/ /0 0 100 0.001 1 1 4/ }' OK test: CRM114 '-{ isolate (:s:); {classify < sks > ( i_test.css | q_test.css | i_vs_q_test.css ) (:s:) /[[:graph:]]+/ /0 0 100 0.001 1 1 4/ [:_dw:] ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v prob=0.0004 -v pR=0.03 OK test: CRM114 '-{ isolate (:s:); {classify < sks > ( i_test.css | q_test.css | i_vs_q_test.css ) (:s:) /[[:graph:]]+/ /0 0 100 0.001 1 1 4/ [:_dw:] ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v prob=2e-4 -v pR=0.04 OK ******************************************* * String Kernel SVM (SKS) Unique classifier ******************************************* rm -f i_vs_q_test.css rm -f i_test.css rm -f q_test.css test: CRM114 '-{ match <fromend> (:one_paragraph:) /[[:graph:]]+.*?\n\n/; translate [:one_paragraph:] (:one_paragraph:) /.,!?@#^&*()/; learn [:one_paragraph:] < sks unique > (i_test.css) /[[:graph:]]+/ / 0 0 100 0.001 1 1 4/; liaf}' OK test: CRM114 '-{ match <fromend> (:one_paragraph:) /[[:graph:]]+.*?\n\n/; translate [:one_paragraph:] (:one_paragraph:) /.,!?@#^&*()/; learn [:one_paragraph:] < sks unique > (q_test.css) /[[:graph:]]+/ /0 0 100 0.001 1 1 4/ ; liaf }' OK build the actual hyperplanes test: CRM114 '-{window; learn ( i_test.css | q_test.css | i_vs_q_test.css ) < sks unique > /[[:graph:]]+/ /0 0 100 0.001 1 1 4/ }' OK test: CRM114 '-{ isolate (:s:); translate /.,!?@#^&*()/; {classify < sks unique > ( i_test.css | q_test.css | i_vs_q_test.css ) (:s:) /[[:graph:]]+/ /0 0 100 0.001 1 1 4/ [:_dw:] ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ } }' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v features=10 -v prob=0.004 -v pR=0.1 OK test: CRM114 '-{ isolate (:s:); translate /.,!?@#^&*()/; {classify < sks unique > ( i_test.css | q_test.css | i_vs_q_test.css ) (:s:) /[[:graph:]]+/ /0 0 100 0.001 1 1 4/ [:_dw:] ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v features=10 -v prob=0.004 -v pR=0.1 OK ******************************************* * Bytewise Correlation classifier ******************************************* rm -f i_vs_q_test.css rm -f i_test.css rm -f q_test.css test: CRM114 '-{ isolate (:s:) {classify <correlate> ( ./INTRO_mt_ng_reference_2.input | ./QUICKREF_mt_ng_reference_1.input ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v features=10 -v prob=0.004 -v pR=0.1 OK test: CRM114 '-{ isolate (:s:) {classify <correlate> ( ./INTRO_mt_ng_reference_2.input | ./QUICKREF_mt_ng_reference_1.input ) (:s:) /[[:graph:]]+/ ; output / type I \n:*:s:\n/} alius { output / type Q \n:*:s:\n/ }}' targetname = ./crm114_classify.filter.sh.temp AWK filter tolerances / args: -v features=10 -v prob=0.004 -v pR=0.1 OK ******************************************* * Clump \/ Pmulc Test ******************************************* rm -f i_test.css rm -f q_test.css test: CRM114 '-{ match <fromend> (:one_paragraph:) /([[:graph:]]+.*?\n\n){5}/; clump <bychunk> [:one_paragraph:] (i_test.css) /[[:graph:]]+/; output /./ ; liaf}' ---- diff report: ---- --- ./ref/Clump_Pmulc_test1.step1.refoutput 2008-05-01 08:39:33.000000000 +0200 +++ ./testscript.sh.tempchk 2008-08-12 10:27:14.000000000 +0200 @@ -2,9 +2,16 @@ -{ match <fromend> (:one_paragraph:) /([[:graph:]]+.*?\n\n){5}/; clump <bychunk> [:one_paragraph:] (i_test.css) /[[:graph:]]+/; output /./ ; liaf} RETURN CODE: -0 +154 STDERR TEST OUTPUT: +crm114: *UNTRAPPABLE ERROR* + Cannot allocate distance matrix M[] (C) Stick a fork in us; we're _done_. +Sorry, but this program is very sick and probably should be killed off. +This happened at line 3 of file (from command line): + clump <bychunk> [:one_paragraph:] (i_test.css) /[[:graph:]]+/ +(runtime system location: crm_expr_clump.c(782) in routine: thresholding_average_cluster) + STDOUT TEST OUTPUT: ---------------------- FAIL(OVERRIDEN:OK) test: CRM114 '-{ match <fromend> (:one_paragraph:) /([[:graph:]]+.*?\n\n){5}/; clump [:one_paragraph:] <bychunk> (i_test.css) /[[:graph:]]+/; output /./; liaf }' ---- diff report: ---- --- ./ref/Clump_Pmulc_test1.step2.refoutput 2008-05-01 08:39:41.000000000 +0200 +++ ./testscript.sh.tempchk 2008-08-12 10:27:14.000000000 +0200 @@ -2,9 +2,18 @@ -{ match <fromend> (:one_paragraph:) /([[:graph:]]+.*?\n\n){5}/; clump [:one_paragraph:] <bychunk> (i_test.css) /[[:graph:]]+/; output /./; liaf } RETURN CODE: -0 +154 STDERR TEST OUTPUT: +crm114: *UNTRAPPABLE ERROR* + +Better start screaming, guv', since the software's just gone critical: +assertion '!(scale <= 1.19209290e-7F && scale >= -1.19209290e-7F)' failed! +Sorry, but this program is very sick and probably should be killed off. +This happened at line 3 of file (from command line): + clump [:one_paragraph:] <bychunk> (i_test.css) /[[:graph:]]+/ +(runtime system location: crm_expr_clump.c(841) in routine: thresholding_average_cluster) + STDOUT TEST OUTPUT: ---------------------- FAIL(OVERRIDEN:OK) Now see where our paragraphs go to test: CRM114 '-{ isolate (:s:); { pmulc ( i_test.css) (:s:) <bychunk> /[[:graph:]]+/ [:_dw:] ; output /Likely result: \n:*:s:\n/} alius { output / Unsure result \n:*:s:\n/ } }' ---- diff report: ---- --- ./ref/Clump_Pmulc_test1.step3.refoutput 2008-05-01 08:40:09.000000000 +0200 +++ ./testscript.sh.tempchk 2008-08-12 10:27:14.000000000 +0200 @@ -7,18 +7,10 @@ STDERR TEST OUTPUT: STDOUT TEST OUTPUT: -...............................................................Likely result: -PMULC succeeds; success probabilty: 1.0000 pR: 61.5681 -Best match to clump #8 (clump_#8) prob: 1.0000 pR: 61.5681 -Closest document: #46 (document_#46) affinity: 0.8233 + Unsure result +PMULC fails; success probability: 0.0000 pR: 0.0000 +Best match to clump #1 (clump_#1) prob: 0.0000 pR: 0.0000 +Closest document: #-1 (document_#-1) affinity: -1.0000 Total features in input file: 259 -1: (clump_#1): documents: 4 affinity: 0.1529 prob: 0.0000 pR: -70.0000 -2: (clump_#2): documents: 2 affinity: 0.0942 prob: 0.0000 pR: -70.0000 -3: (clump_#3): documents: 49 affinity: 0.1234 prob: 0.0000 pR: -70.0000 -4: (clump_#4): documents: 3 affinity: 0.1136 prob: 0.0000 pR: -70.0000 -5: (clump_#5): documents: 1 affinity: 0.1408 prob: 0.0000 pR: -70.0000 -6: (clump_#6): documents: 1 affinity: 0.1379 prob: 0.0000 pR: -70.0000 -7: (clump_#7): documents: 1 affinity: 0.0991 prob: 0.0000 pR: -70.0000 -8: (clump_#8): documents: 2 affinity: 0.5186 prob: 1.0000 pR: 61.5681 ---------------------- FAIL(OVERRIDEN:OK) test: CRM114 '-{ isolate (:s:); { pmulc ( i_test.css) (:s:) <bychunk> /[[:graph:]]+/ [:_dw:] ; output /Likely result: \n:*:s:\n/} alius { output / Unsure result \n:*:s:\n/ } }' ---- diff report: ---- --- ./ref/Clump_Pmulc_test1.step4.refoutput 2008-05-01 08:40:23.000000000 +0200 +++ ./testscript.sh.tempchk 2008-08-12 10:27:14.000000000 +0200 @@ -7,18 +7,10 @@ STDERR TEST OUTPUT: STDOUT TEST OUTPUT: -Likely result: -PMULC succeeds; success probabilty: 1.0000 pR: 59.1777 -Best match to clump #2 (clump_#2) prob: 1.0000 pR: 59.1777 -Closest document: #1 (document_#1) affinity: 0.7596 + Unsure result +PMULC fails; success probability: 0.0000 pR: 0.0000 +Best match to clump #1 (clump_#1) prob: 0.0000 pR: 0.0000 +Closest document: #-1 (document_#-1) affinity: -1.0000 Total features in input file: 223 -1: (clump_#1): documents: 4 affinity: 0.1356 prob: 0.0000 pR: -70.0000 -2: (clump_#2): documents: 2 affinity: 0.4753 prob: 1.0000 pR: 59.1777 -3: (clump_#3): documents: 49 affinity: 0.1422 prob: 0.0000 pR: -70.0000 -4: (clump_#4): documents: 3 affinity: 0.1230 prob: 0.0000 pR: -70.0000 -5: (clump_#5): documents: 1 affinity: 0.1208 prob: 0.0000 pR: -70.0000 -6: (clump_#6): documents: 1 affinity: 0.1421 prob: 0.0000 pR: -70.0000 -7: (clump_#7): documents: 1 affinity: 0.1173 prob: 0.0000 pR: -70.0000 -8: (clump_#8): documents: 2 affinity: 0.0928 prob: 0.0000 pR: -70.0000 ---------------------- FAIL(OVERRIDEN:OK) test: CRM114 '-{ isolate (:s:); { pmulc ( i_test.css) (:s:) <bychunk> /[[:graph:]]+/ [:_dw:] ; output /Likely result: \n:*:s:\n/} alius { output / Unsure result \n:*:s:\n/ } }' ---- diff report: ---- --- ./ref/Clump_Pmulc_test1.step5.refoutput 2008-05-01 08:40:39.000000000 +0200 +++ ./testscript.sh.tempchk 2008-08-12 10:27:14.000000000 +0200 @@ -8,68 +8,7 @@ STDOUT TEST OUTPUT: Likely result: -0 (document_#0) clump: 1 (clump_#1) affinity: 0.1439 -1 (document_#1) clump: 2 (clump_#2) affinity: 0.0887 -2 (document_#2) clump: 3 (clump_#3) affinity: 0.1784 -3 (document_#3) clump: 4 (clump_#4) affinity: 0.1651 -4 (document_#4) clump: 4 (clump_#4) affinity: 0.1949 -5 (document_#5) clump: 4 (clump_#4) affinity: 0.2162 -6 (document_#6) clump: 3 (clump_#3) affinity: 0.1953 -7 (document_#7) clump: 1 (clump_#1) affinity: 0.1418 -8 (document_#8) clump: 1 (clump_#1) affinity: 0.1535 -9 (document_#9) clump: 1 (clump_#1) affinity: 0.1432 -10 (document_#10) clump: 3 (clump_#3) affinity: 0.1790 -11 (document_#11) clump: 3 (clump_#3) affinity: 0.1883 -12 (document_#12) clump: 3 (clump_#3) affinity: 0.2052 -13 (document_#13) clump: 3 (clump_#3) affinity: 0.1841 -14 (document_#14) clump: 3 (clump_#3) affinity: 0.2289 -15 (document_#15) clump: 3 (clump_#3) affinity: 0.1805 -16 (document_#16) clump: 3 (clump_#3) affinity: 0.1776 -17 (document_#17) clump: 3 (clump_#3) affinity: 0.1966 -18 (document_#18) clump: 2 (clump_#2) affinity: 0.0887 -19 (document_#19) clump: 3 (clump_#3) affinity: 0.1803 -20 (document_#20) clump: 3 (clump_#3) affinity: 0.2188 -21 (document_#21) clump: 3 (clump_#3) affinity: 0.2216 -22 (document_#22) clump: 3 (clump_#3) affinity: 0.2264 -23 (document_#23) clump: 3 (clump_#3) affinity: 0.1826 -24 (document_#24) clump: 3 (clump_#3) affinity: 0.2031 -25 (document_#25) clump: 3 (clump_#3) affinity: 0.1978 -26 (document_#26) clump: 3 (clump_#3) affinity: 0.1834 -27 (document_#27) clump: 3 (clump_#3) affinity: 0.2121 -28 (document_#28) clump: 3 (clump_#3) affinity: 0.1853 -29 (document_#29) clump: 3 (clump_#3) affinity: 0.1602 -30 (document_#30) clump: 3 (clump_#3) affinity: 0.1999 -31 (document_#31) clump: 3 (clump_#3) affinity: 0.2193 -32 (document_#32) clump: 3 (clump_#3) affinity: 0.2209 -33 (document_#33) clump: 3 (clump_#3) affinity: 0.1977 -34 (document_#34) clump: 5 (clump_#5) affinity: 0.0000 -35 (document_#35) clump: 6 (clump_#6) affinity: 0.0000 -36 (document_#36) clump: 3 (clump_#3) affinity: 0.1724 -37 (document_#37) clump: 3 (clump_#3) affinity: 0.1824 -38 (document_#38) clump: 3 (clump_#3) affinity: 0.1955 -39 (document_#39) clump: 3 (clump_#3) affinity: 0.1970 -40 (document_#40) clump: 3 (clump_#3) affinity: 0.1906 -41 (document_#41) clump: 3 (clump_#3) affinity: 0.1979 -42 (document_#42) clump: 3 (clump_#3) affinity: 0.1926 -43 (document_#43) clump: 3 (clump_#3) affinity: 0.1717 -44 (document_#44) clump: 7 (clump_#7) affinity: 0.0000 -45 (document_#45) clump: 3 (clump_#3) affinity: 0.1876 -46 (document_#46) clump: 8 (clump_#8) affinity: 0.1011 -47 (document_#47) clump: 8 (clump_#8) affinity: 0.1011 -48 (document_#48) clump: 3 (clump_#3) affinity: 0.1751 -49 (document_#49) clump: 3 (clump_#3) affinity: 0.1998 -50 (document_#50) clump: 3 (clump_#3) affinity: 0.1896 -51 (document_#51) clump: 3 (clump_#3) affinity: 0.1863 -52 (document_#52) clump: 3 (clump_#3) affinity: 0.2194 -53 (document_#53) clump: 3 (clump_#3) affinity: 0.2076 -54 (document_#54) clump: 3 (clump_#3) affinity: 0.2291 -55 (document_#55) clump: 3 (clump_#3) affinity: 0.2257 -56 (document_#56) clump: 3 (clump_#3) affinity: 0.2070 -57 (document_#57) clump: 3 (clump_#3) affinity: 0.1970 -58 (document_#58) clump: 3 (clump_#3) affinity: 0.2020 -59 (document_#59) clump: 3 (clump_#3) affinity: 0.2016 -60 (document_#60) clump: 3 (clump_#3) affinity: 0.1767 -61 (document_#61) clump: 3 (clump_#3) affinity: 0.1724 -62 (document_#62) clump: 3 (clump_#3) affinity: 0.1727 +0 (document_#0) clump: 0 (unassigned) affinity: 0.0822 +1 (document_#1) clump: 0 (unassigned) affinity: 0.0822 ---------------------- FAIL(OVERRIDEN:OK) ******************************************* * TESTING DONE! * All 'megatest'-derived test cases have * been checked: OK (some may have been * 'OVERRIDDEN', please check the output * when you worry about those... * ******************************************* make[2]: Leaving directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/tests' make[2]: Entering directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/tests' Makefile:1908: warning: overriding commands for target `check-am' Makefile:847: warning: ignoring old commands for target `check-am' rm -f i_test.css rm -f i_vs_q_test.css rm -f q_test.css rm -f randtst.txt rm -rf *.temp.* rm -rf *.temperr rm -rf *.tempchk rm -rf *.diffout rm -rf *.tempout rm -rf *.filtout make[2]: Leaving directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/tests' make[1]: Leaving directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason/tests' make[1]: Entering directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason' make[1]: Nothing to be done for `check-am'. make[1]: Leaving directory `/home/tsp/programs/crm114/crm114-20080502-BlameJason' > > > Thank you all for your excellent work and I hope this report may be > > useful. > > Absolutely: very useful! Thanks again! > > Thank you for your detailed comments on the megatest diffs. I will study these carefully. My best regards Thomas Spahni ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/