diff options
author | Hieu Hoang <hieu@hoang.co.uk> | 2013-06-26 20:14:10 +0400 |
---|---|---|
committer | Hieu Hoang <hieu@hoang.co.uk> | 2013-06-26 20:14:10 +0400 |
commit | f13d5919145e209fb3db9424b9e7920730b6800e (patch) | |
tree | 3a23b83cd628503c24f8c3ed34a5c666c425e226 /tests | |
parent | 5d3870a68fc373bec6f63161ba7246510bf76ac5 (diff) | |
parent | 2cbf31ca9f95d5b99750fe50e2c3ad5fa80d326e (diff) |
Merge branch 'master' of github.com:moses-smt/moses-regression-tests
Diffstat (limited to 'tests')
18 files changed, 207 insertions, 0 deletions
diff --git a/tests/chart.suffix-array/filter-stderr.pl b/tests/chart.suffix-array/filter-stderr.pl new file mode 100755 index 0000000..2f6e176 --- /dev/null +++ b/tests/chart.suffix-array/filter-stderr.pl @@ -0,0 +1,22 @@ +#!/usr/bin/perl + +BEGIN { use Cwd qw/ abs_path /; use File::Basename; $script_dir = dirname(abs_path($0)); push @INC, "$script_dir/../perllib"; } +use RegTestUtils; + +$x=0; +while (<>) { + chomp; + + if (/^Finished loading LanguageModels/) { + my $time = RegTestUtils::readTime($_); + print "LMLOAD_TIME ~ $time\n"; + } + if (/^Finished loading phrase tables/) { + my $time = RegTestUtils::readTime($_); + print "PTLOAD_TIME ~ $time\n"; + } + next unless /^BEST TRANSLATION:/; + my $pscore = RegTestUtils::readHypoScore($_); + $x++; + print "SCORE_$x = $pscore\n"; +} diff --git a/tests/chart.suffix-array/filter-stdout.pl b/tests/chart.suffix-array/filter-stdout.pl new file mode 100755 index 0000000..476ddf6 --- /dev/null +++ b/tests/chart.suffix-array/filter-stdout.pl @@ -0,0 +1,7 @@ +#!/usr/bin/perl +$x=0; +while (<>) { + chomp; + $x++; + print "TRANSLATION_$x=$_\n"; +} diff --git a/tests/chart.suffix-array/glue-grammar.1 b/tests/chart.suffix-array/glue-grammar.1 new file mode 100644 index 0000000..c17513c --- /dev/null +++ b/tests/chart.suffix-array/glue-grammar.1 @@ -0,0 +1,3 @@ +<s> [X] ||| <s> [S] ||| 1 ||| ||| 0 +[X][S] </s> [X] ||| [X][S] </s> [S] ||| 1 ||| 0-0 ||| 0 +[X][S] [X][X] [X] ||| [X][S] [X][X] [S] ||| 2.718 ||| 0-0 1-1 ||| 0 diff --git a/tests/chart.suffix-array/grammar.0.gz b/tests/chart.suffix-array/grammar.0.gz Binary files differnew file mode 100644 index 0000000..ecad55d --- /dev/null +++ b/tests/chart.suffix-array/grammar.0.gz diff --git a/tests/chart.suffix-array/grammar.1.gz b/tests/chart.suffix-array/grammar.1.gz Binary files differnew file mode 100644 index 0000000..0559067 --- /dev/null +++ b/tests/chart.suffix-array/grammar.1.gz diff --git a/tests/chart.suffix-array/grammar.2.gz b/tests/chart.suffix-array/grammar.2.gz Binary files differnew file mode 100644 index 0000000..afa3d1e --- /dev/null +++ b/tests/chart.suffix-array/grammar.2.gz diff --git a/tests/chart.suffix-array/grammar.3.gz b/tests/chart.suffix-array/grammar.3.gz Binary files differnew file mode 100644 index 0000000..efc4ec7 --- /dev/null +++ b/tests/chart.suffix-array/grammar.3.gz diff --git a/tests/chart.suffix-array/grammar.4.gz b/tests/chart.suffix-array/grammar.4.gz Binary files differnew file mode 100644 index 0000000..48e5007 --- /dev/null +++ b/tests/chart.suffix-array/grammar.4.gz diff --git a/tests/chart.suffix-array/grammar.5.gz b/tests/chart.suffix-array/grammar.5.gz Binary files differnew file mode 100644 index 0000000..c62744f --- /dev/null +++ b/tests/chart.suffix-array/grammar.5.gz diff --git a/tests/chart.suffix-array/grammar.6.gz b/tests/chart.suffix-array/grammar.6.gz Binary files differnew file mode 100644 index 0000000..294075c --- /dev/null +++ b/tests/chart.suffix-array/grammar.6.gz diff --git a/tests/chart.suffix-array/grammar.7.gz b/tests/chart.suffix-array/grammar.7.gz Binary files differnew file mode 100644 index 0000000..ae42f4b --- /dev/null +++ b/tests/chart.suffix-array/grammar.7.gz diff --git a/tests/chart.suffix-array/grammar.8.gz b/tests/chart.suffix-array/grammar.8.gz Binary files differnew file mode 100644 index 0000000..edfdecf --- /dev/null +++ b/tests/chart.suffix-array/grammar.8.gz diff --git a/tests/chart.suffix-array/grammar.9.gz b/tests/chart.suffix-array/grammar.9.gz Binary files differnew file mode 100644 index 0000000..368666d --- /dev/null +++ b/tests/chart.suffix-array/grammar.9.gz diff --git a/tests/chart.suffix-array/input.sgm b/tests/chart.suffix-array/input.sgm new file mode 100644 index 0000000..a5a09e6 --- /dev/null +++ b/tests/chart.suffix-array/input.sgm @@ -0,0 +1,100 @@ +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.0.gz" id="0">Barack Obama erhält als vierter US @-@ Präsident den Friedensnobelpreis</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.1.gz" id="1">der amerikanische Präsident Barack Obama kommt für 26 Stunden nach Oslo , Norwegen , um hier als vierter US @-@ Präsident in der Geschichte den Friedensnobelpreis entgegenzunehmen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.2.gz" id="2">darüber hinaus erhält er das Diplom sowie die Medaille und einen Scheck über 1,4 Mio. Dollar für seine außergewöhnlichen Bestrebungen um die Intensivierung der Weltdiplomatie und Zusammenarbeit unter den Völkern .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.3.gz" id="3">der Chef des Weißen Hauses kommt morgen zusammen mit seiner Frau Michelle in der nordwegischen Metropole an und wird die ganze Zeit beschäftigt sein .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.4.gz" id="4">zunächst stattet er dem Nobel @-@ Institut einen Besuch ab , wo er überhaupt zum erstenmal mit den fünf Ausschussmitglieder zusammen trifft , die ihn im Oktober aus 172 Leuten und 33 Organisationen gewählt haben .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.5.gz" id="5">das Präsidenten paar hat danach ein Treffen mit dem norwegischen König Harald V. und Königin Sonja eingeplant .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.6.gz" id="6">nachmittags erreicht dann der Besuch seinen Höhe Punkt mit der Zeremonie , bei der Obama den Prestigepreis übernimmt .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.7.gz" id="7">diesen erhält er als der vierte US @-@ Präsident , aber erst als der dritte , der den Preis direkt im Amt entgegen nimmt .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.8.gz" id="8">das Weiße Haus avisierte schon , dass Obama bei der Übernahme des Preises über den Afghanistankrieg sprechen wird .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.9.gz" id="9">der Präsident will diesem Thema nicht ausweichen , weil er weiß , dass er den Preis als ein Präsident übernimmt , der zur Zeit Krieg in zwei Ländern führt .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.10.gz" id="10">die Kriegsanstrengungen hat er vor einigen Tagen noch durch Aussendung weiterer Truppen nach Afghanistan verstärkt , was keiner von seinen Kritikern vergisst , hervorzuheben .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.11.gz" id="11">bei der Zeremonie erhielt Obama auch die Golmedaille , das Diplom und den Scheck über zehn Millionen schwedischen Kronen ( ca. 24 Millionen Tschechischen Kronen ) .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.12.gz" id="12">das Geld beabsichtigt er , für wohltätige Zwecke zu spenden , aber über ein konkretes karitatives Werk hat er noch nicht entschieden .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.13.gz" id="13">das Veranstaltungsprogramm sieht dann ein Bankett mit dem Präsidenten und seiner Frau vor , bei dem außer Norwegens König und Königin auch der Premierminister und weitere 250 geladene Gäste anwesend sein werden .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.14.gz" id="14">Obama war von Anfang dem Preis gegenüber sehr zurückhaltend .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.15.gz" id="15">er führte Beispiel Weise an , dass er das Gefühl habe , dass er den Preis nicht ganz verdiene .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.16.gz" id="16">mehrmals wiederholte er auch , dass es sich um den Preis nicht nur für ihn handelt , sondern für alle , die die gleichen Werte wie er anstreben .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.17.gz" id="17">sofort nach der Bekanntgabe führte er auf , dass das für ihn " Aufforderung zur Handlung " ist .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.18.gz" id="18">Naturschützer wird der Erpressung beschuldigt</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.19.gz" id="19">die Leitmeritz @-@ Polizei beschuldigte den Vorsitzenden der Bürger Vereinigung " Natur Schutz Gemeinschaft Leitmeritz " wegen Erpressung .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.20.gz" id="20">in mehreren Fällen legte er in letzten Jahr Berufungen gegen einige Bauverfahren ein , für die Rücknahme seiner Berufungen forderte er von den Investoren Geld zu seinen Gunsten , sagte dieLeitmeritz @-@ Polizeisprecherin , Alena Romova .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.21.gz" id="21">der Vorsitzende der Leitmeritz @-@ Naturschutzgemeinschaftist Lubomir Studnicka .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.22.gz" id="22">jetzt ist er in Haft und es droht ihm eine Gefängnisstrafe von bis zu drei Jahren .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.23.gz" id="23">die Sortier @-@ Müllcontainer in Brünn riechen nicht gut</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.24.gz" id="24">in Prag können die Menschen die Getränkeverpackungen sortieren , in den südmährischen Ortschaften findet man wieder auf Schritt und tritt die Sortier @-@ Müllcontainer .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.25.gz" id="25">und in Brünn ?</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.26.gz" id="26">dort müsste man ein paar Minuten reservieren , um die geeigneten Mülltonnen zu finden .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.27.gz" id="27">ich will nicht behaupten , dass die Abfalltrennung in Brünn überhaupt nicht funktionieren würde .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.28.gz" id="28">aber es kommt mir vor , al ob die Rats Herren sich um dieses Problem unzureichend kümmern , auch wenn wir die zweitgrößte Stadt in der Republik sind .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.29.gz" id="29">Brünn befindet sich in der Abfalltrennen in der Urwaldliga , nicht nur weil wir als Bürger hier in die Sortiercontainer immer nur Papier , Glas und Kunststoffe entsorgen dürfen , aber vor allem durch die Mülltonnenanzahl für solche Abfälle .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.30.gz" id="30">haben sie irgendwann im Zentrum versucht , zum Beispiel die PET @-@ Flaschen in einem Sammelkorb zu entsorgen ?</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.31.gz" id="31">das ist eben Kunst , vor allem braucht es Zeit und starke Nerven .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.32.gz" id="32">die einzige Stelle , die ich bis jetzt zufällig gefunden habe , ist der Mährenplatz gegenüber der getroffenen Gans .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.33.gz" id="33">Herr Onderka uns seine Kollegen sollen sich nicht ärgern , aber das kommt mir viel zu wenig vor .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.34.gz" id="34">abgesehen davon , dass die Abfalltrennung auch in anderen Teilen kein Honigschlecken ist .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.35.gz" id="35">ganze Reihen meiner Freunde sortieren deshalb den Abfall lieber gar nicht .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.36.gz" id="36">aus Faulheit .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.37.gz" id="37">es ist ihnen nämlich zu weit zu den Mülltonnen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.38.gz" id="38">ich wundere mich nicht .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.39.gz" id="39">ich habe selbst nach der Feier angeboten , ein paar Glas- und Plastikflaschen zu entsorgen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.40.gz" id="40">aber auf der Konicova Strasse habe ich in Sichtweite keine Farbmülltonnen gesehen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.41.gz" id="41">zum Glück kam ich beim Straßenbahnfahren an die richtige Stelle .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.42.gz" id="42">der Müll lief aber fast über .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.43.gz" id="43">das ist doch immer noch das Zentrum von Brünn , wodurch die ganze Stadt repräsentiert wird , nicht wahr ? !</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.44.gz" id="44">allerdings lässt Ökologie und irgend eine Art von Ästhetik die Brünner Sozialisten wahrscheinlich kalt .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.45.gz" id="45">sie kaufen lieber der Stadt eine Eishockeyliga , die uns in der ganzen Republik Schande macht , anstatt ein paar Mülltonnen mehr zu kaufen und die Sortierung weiterer Rohstoffe für ein saubereres Brünn anzubieten .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.46.gz" id="46">ich bin froh , dass ich meine Annahme auch mit offiziellen Statistiken bestätigen konnte , nachdem ich eine Weile im Internet gesucht habe .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.47.gz" id="47">nach den Daten der Gesellschaft EKO @-@ KOM steht Brünn wirklich in ganzem Bezirk Südmähren am schlimmsten da .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.48.gz" id="48">das " Paroubek " -Budget hat das Geld für Renten und Krankengeld entzogen</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.49.gz" id="49">nachdem die Linkspartei trotz der Ansicht der Regierung den Landwirten und Beamten mehr Geld gab , ist im Budget kein Geld für Renten , Krankengeld und Bauspargeschäft mehr vorhanden .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.50.gz" id="50">aber auch nicht für die Staatsschuldzinsen oder für die internationale Gerichtsstreite .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.51.gz" id="51">das verursacht denen noch mehr Probleme , die in der zweiten Hälfte nächstes Jahres an der Regierung sein werden , sagt der Finanzminister , Eduard Janota .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.52.gz" id="52">als die Sozialdemokraten und Kommunisten die Gehaltserhöhung für einige Gruppen durchgesetzt haben , rechneten sie damit , dass die Regierung sich das Geld für die empfindlichen Ausgaben , Altersrenten oder für das Krankengeld zusätzlich borgt . " das ist nur ein Aufschub der Probleme in die Zukunft , Minenlegen , " sagt Janota .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.53.gz" id="53">von den Krediten werden nähmlich immer höhere Zinsen bezahlt , und man muss das Geld irgendwann zurückerstatten .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.54.gz" id="54">zuviel zur Auswahl , woher nehmen , die Staat Kasse ist leer .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.55.gz" id="55">falls die Sozialleistungen oder Kosten für den Staat Betrieb nicht gekürzt werden , sind die Steuer zu erhöhen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.56.gz" id="56">für all die , denen die Linkspartei jetzt mehr Geld gab : der Feuerwehr , den Lehrern , den Landwirten .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.57.gz" id="57">und später auch für ihre Kinder .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.58.gz" id="58">um die Renten oder das Krankengeld , von deren Konto 1,8 Milliarden verschwunden sind , müssen die Leute keine Angst haben , sie bekommen dieses Geld .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.59.gz" id="59">aber um den Preis , dass der Staat sich noch mehr verschuldet - im nächstem Jahr werden so in der Staat Kasse nicht nur 163 Milliarden fehlen , sondern weit mehr .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.60.gz" id="60">trotz den Bemühungen des Ministers Janota wird die Verschuldung so immer schneller steigen und zusammen mit dieser auch die Kosten und Zinsen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.61.gz" id="61">der Staat wird das Geld für den Ausgleich brauchen , dann wird schon nicht mehr ausreichen die mehr Wert Steuer nur geringfügig zu erhöhen oder das Mutterschaftsgeld zu kürzen , wie das die Regierung in ihrem Gegendefizit @-@ Paket machte .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.62.gz" id="62">wegen der Geldabwälzung zugunsten einiger Wahl Volk Gruppen werden in nächstem Jahr auch fünf Milliarden für Ausbau der Autobahnen und Eisenbahnen fehlen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.63.gz" id="63">es handelt sich überwiegend um die im Wettbewerb gewonnenen Abschnitte , die man nicht mehr neu projektieren und eventuell verbilligen kann .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.64.gz" id="64">aller höchstens erhalten , was auch nicht kostenlos ist .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.65.gz" id="65">das bedeutet Verzug oder Einstellung von dreißig großen Bauwerken , gefährdet ist auch die Europa @-@ Geldschöpfung , warnt der Sprecher des Verkehrsministeriums , Karel Hanzelka .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.66.gz" id="66">die CSSD hat vorgeschlagen , dass sie den Abfluss von Geld salomonisch löst - sie greift die CEZ @-@ Aktiendividenden an , bei denen der Staat Eigentümer der Majorität ist , und aus diesen dann das Geld in den Verkehr Fonds ausgeschüttet werden .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.67.gz" id="67">das Problem besteht darin , dass die Höhe der Dividende niemals sicher ist und es handelt sich keinesfalls um eine unendliche Quelle .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.68.gz" id="68">im vorigen Jahr flossen aus CEZ 18 Milliarden ins Budget .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.69.gz" id="69">Alljährlich werden aus den Dividenden die Umwelt Schäden @-@ Liquidationenbezahlt , das Pensionskonto aufgefüllt und dieses Geld dient auch als Reserve für die geplante Rentenreform .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.70.gz" id="70">mit diesem Trick steigt darüber hinaus nach den Regeln der EU das tatsächliche Budgetdefizit auf das Nieveau von 5,7 Prozent BIP .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.71.gz" id="71">damit endet aber der Karneval der Umwälzungen noch nicht .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.72.gz" id="72">fast drei Milliarden sank das Budget für verschiedene Entschädigungen für Opfer des Kommunismus und von Straftaten oder für Aufwendungen für die Gerichtsstreite , sogar für das Bausparen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.73.gz" id="73">ob der Staat ohne dieses Geld zurechtkommt , ist nicht sicher .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.74.gz" id="74">das ist reines Glücksspiel .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.75.gz" id="75">wenn wir einen internationalen Gerichtsstreit verlieren , dann müssen wir das sowieso bezahlen , stellt Janota klar .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.76.gz" id="76">der Minister muss dann das Geld Ressort intern einsparen oder im schlimmsten Fall die Regierungsbudgetreserve angreifen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.77.gz" id="77">einhundert Millionen fehlen auch für das schon angefangene zwei Milliarden Projekt " Staat Kasse " , das ermöglichen soll , on @-@ Line zu überwachen , für was die Einzelbehörden das Geld ausgeben .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.78.gz" id="78">die erste Etappe startet im Januar .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.79.gz" id="79">sollte der Finanzminister das Geld nicht anderswo finden , müsste das Projekt gestoppt werden und in diesem Falle kommen Sanktionen , warnte Janota .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.80.gz" id="80">ihr nächstes Smartphone wird zwei Betrieb Systeme beherrschen</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.81.gz" id="81">die Amerikaner rechnen für die Zukunft mit einem Handy , auf dem der Benutzer durch Drücken einer einzigen Taste zwischen verschiedenen Betriebssystemen umschalten kann .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.82.gz" id="82">die vorgelegten Pläne sehen vielversprechend aus .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.83.gz" id="83">es reicht , nur eine Taste zu drücken , und aus Windows Mobile wird innerhalb weniger Sekunden im Wechsel Android .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.84.gz" id="84">genau diese Möglichkeit will die amerikanische Gesellschaft VMware vorstellen , die sich vor allem auf die Entwicklung des virtuellen Softwares für die Computer konzentriert .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.85.gz" id="85">wir ermöglichen Ihnen , auf dem Handy zwei Anwenderprofile gleichzeitig zu auszuführen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.86.gz" id="86">sie können zwischen diesen umschalten , sie können diese in zwei Home- und Arbeitsprofile trennen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.87.gz" id="87">beide werden parallel arbeiten , deutete Srinivas Krishnamurti von der Firma VMware beim Interview für das Magazin Computerworld an .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.88.gz" id="88">es wurde im November letzten Jahres präsentiert und vor einigen Tagen zum ersten Mal vorgeführt .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.89.gz" id="89">auf dem Markt wird es 2012 kommen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.90.gz" id="90">die Virtualisierung der Smart @-@ Phones ist keine Fiktion .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.91.gz" id="91">VMware führte den Journalisten schon das Smartphone @-@ Testmuster mit zwei Betriebssystemen erfolgreich vor .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.92.gz" id="92">eines war ein modifiziertes Nokia N800 mit einem RAM @-@ Speicher von 128 MB , auf dem parallel gleichzeitig die erwähnten Systeme Windows Mobile und Android liefen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.93.gz" id="93">die Entwicklung der neuen Technologie für die Mobiltelefone ist in vollem Zuge .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.94.gz" id="94">jetzt arbeitet die Firma VMware zusammen mit den europäischen und amerikanischen Betreibern an der Smartphonen @-@ Virtualisirung und zum Kunden sollen die Geräte innerhalb des Jahres 2012 kommen .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.95.gz" id="95">Schneemangel auf den Bergen quält die Hotelbesitzer</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.96.gz" id="96">Scneelose Pisten machen Sorgenn nicht nur den Riesengebirgen @-@ Anwohnern .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.97.gz" id="97">der fehlende Schnee schreckt die Leute auch von Bestellung der Skiaufenthalte in Hotels und Pensionen ab .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.98.gz" id="98">deshalb kann man in Riesengebirge immer noch freie Zimmer für alle Wintertermine incl . Weihnachten und Silvester erhalten .</seg> +<seg grammar="/home/waziz/experiments/xrce/toy/tuning/filtered.1/grammar.99.gz" id="99">unsere Seiten werden häufig besucht .</seg> diff --git a/tests/chart.suffix-array/interpolated-binlm.1 b/tests/chart.suffix-array/interpolated-binlm.1 Binary files differnew file mode 100644 index 0000000..b3f61f0 --- /dev/null +++ b/tests/chart.suffix-array/interpolated-binlm.1 diff --git a/tests/chart.suffix-array/moses.ini b/tests/chart.suffix-array/moses.ini new file mode 100644 index 0000000..e1393ac --- /dev/null +++ b/tests/chart.suffix-array/moses.ini @@ -0,0 +1,44 @@ +######################### +### MOSES CONFIG FILE ### +######################### + +# input factors +[input-factors] +0 + +# mapping steps +[mapping] +0 T 0 +1 T 1 + +[cube-pruning-pop-limit] +1000 + +[non-terminals] +X + +[search-algorithm] +3 + +[inputtype] +3 + +[max-chart-span] +20 +1000 + +# feature functions +[feature] +UnknownWordPenalty +WordPenalty +PhraseDictionaryALSuffixArray name=TranslationModel0 table-limit=20 num-features=7 path=${TEST_PATH} input-factor=0 output-factor=0 +PhraseDictionaryMemory name=TranslationModel1 num-features=1 path=${TEST_PATH}/glue-grammar.1 input-factor=0 output-factor=0 +KENLM lazyken=0 name=LM0 factor=0 path=${TEST_PATH}/interpolated-binlm.1 order=3 + +# dense weights for feature functions +[weight] +UnknownWordPenalty0= 1 +WordPenalty0= -1 +TranslationModel0= 0.2 0.2 0.2 0.2 0.2 0.2 0.2 +TranslationModel1= 1.0 +LM0= 0.5 diff --git a/tests/chart.suffix-array/to-translate.txt b/tests/chart.suffix-array/to-translate.txt new file mode 100644 index 0000000..0aae1af --- /dev/null +++ b/tests/chart.suffix-array/to-translate.txt @@ -0,0 +1,10 @@ +Barack Obama erhält als vierter US @-@ Präsident den Friedensnobelpreis +der amerikanische Präsident Barack Obama kommt für 26 Stunden nach Oslo , Norwegen , um hier als vierter US @-@ Präsident in der Geschichte den Friedensnobelpreis entgegenzunehmen . +darüber hinaus erhält er das Diplom sowie die Medaille und einen Scheck über 1,4 Mio. Dollar für seine außergewöhnlichen Bestrebungen um die Intensivierung der Weltdiplomatie und Zusammenarbeit unter den Völkern . +der Chef des Weißen Hauses kommt morgen zusammen mit seiner Frau Michelle in der nordwegischen Metropole an und wird die ganze Zeit beschäftigt sein . +zunächst stattet er dem Nobel @-@ Institut einen Besuch ab , wo er überhaupt zum erstenmal mit den fünf Ausschussmitglieder zusammen trifft , die ihn im Oktober aus 172 Leuten und 33 Organisationen gewählt haben . +das Präsidenten paar hat danach ein Treffen mit dem norwegischen König Harald V. und Königin Sonja eingeplant . +nachmittags erreicht dann der Besuch seinen Höhe Punkt mit der Zeremonie , bei der Obama den Prestigepreis übernimmt . +diesen erhält er als der vierte US @-@ Präsident , aber erst als der dritte , der den Preis direkt im Amt entgegen nimmt . +das Weiße Haus avisierte schon , dass Obama bei der Übernahme des Preises über den Afghanistankrieg sprechen wird . +der Präsident will diesem Thema nicht ausweichen , weil er weiß , dass er den Preis als ein Präsident übernimmt , der zur Zeit Krieg in zwei Ländern führt . diff --git a/tests/chart.suffix-array/truth/results.txt b/tests/chart.suffix-array/truth/results.txt new file mode 100644 index 0000000..5c587e9 --- /dev/null +++ b/tests/chart.suffix-array/truth/results.txt @@ -0,0 +1,21 @@ +TRANSLATION_1=Barack Obama budget as vierter US President of the Friedensnobelpreis +TRANSLATION_2=the fancy President Barack Obama has on 26 hours the Oslo Agreement , Norway , there is a vierter US President @-@ in the history of the Friedensnobelpreis entgegenzunehmen . +TRANSLATION_3=in addition to budget to the Diplom , as well as the Medaille and a Scheck , 1,4 USD 1 million for abnormal efforts to the Intensivierung Weltdiplomatie and cooperation in the peoples . +TRANSLATION_4=in the of Weißen of it will , in this , the Michelle nordwegischen Metropole , and the whole of , . +TRANSLATION_5=firstly stattet him the Nobel Institut on the on , it is the first five Ausschussmitglieder , which is , in October , 172 Leuten and 33 and elected . +TRANSLATION_6=that is , few , then a meeting with the norwegischen König Harald fifth and Königin Sonja eingeplant . +TRANSLATION_7=nachmittags the will of the to the point of the Zeremonie , also in the Obama Prestigepreis . +TRANSLATION_8=this budget to be the fourth , the US President , it will be the third , the price directly in the office of the . +TRANSLATION_9=the Weiße in avisierte , and dass Obama of Übernahme Preises of the Afghanistankrieg of . +TRANSLATION_10=the President to this does not ausweichen , because they know dass to the price of a President takes the in the war in the two countries . +SCORE_1 = -512.916 +SCORE_2 = -632.887 +SCORE_3 = -632.552 +SCORE_4 = -419.565 +SCORE_5 = -636.810 +SCORE_6 = -612.041 +SCORE_7 = -421.228 +SCORE_8 = -132.327 +SCORE_9 = -711.402 +SCORE_10 = -232.632 +TOTAL_WALLTIME ~ 31 |