Author: rwesten
Date: Thu May 24 14:27:40 2012
New Revision: 1342285
URL: http://svn.apache.org/viewvc?rev=1342285&view=rev
Log:
STANBOL-628: updated the ant files that existing perceptron POS models are
deleted and the maxent versions are downloaded and included in the openNLP
language bundles.
Modified:
incubator/stanbol/trunk/data/opennlp/lang/da/download_models.xml
incubator/stanbol/trunk/data/opennlp/lang/da/src/main/resources/org/apache/stanbol/data/opennlp/
(props changed)
incubator/stanbol/trunk/data/opennlp/lang/de/download_models.xml
incubator/stanbol/trunk/data/opennlp/lang/de/src/main/resources/org/apache/stanbol/data/opennlp/
(props changed)
incubator/stanbol/trunk/data/opennlp/lang/en/download_models.xml
incubator/stanbol/trunk/data/opennlp/lang/nl/download_models.xml
incubator/stanbol/trunk/data/opennlp/lang/nl/src/main/resources/org/apache/stanbol/data/opennlp/
(props changed)
incubator/stanbol/trunk/data/opennlp/lang/pt/download_models.xml
incubator/stanbol/trunk/data/opennlp/lang/pt/src/main/resources/org/apache/stanbol/data/opennlp/
(props changed)
incubator/stanbol/trunk/data/opennlp/lang/sv/download_models.xml
incubator/stanbol/trunk/data/opennlp/lang/sv/src/main/resources/org/apache/stanbol/data/opennlp/
(props changed)
Modified: incubator/stanbol/trunk/data/opennlp/lang/da/download_models.xml
URL:
http://svn.apache.org/viewvc/incubator/stanbol/trunk/data/opennlp/lang/da/download_models.xml?rev=1342285&r1=1342284&r2=1342285&view=diff
==============================================================================
--- incubator/stanbol/trunk/data/opennlp/lang/da/download_models.xml (original)
+++ incubator/stanbol/trunk/data/opennlp/lang/da/download_models.xml Thu May 24
14:27:40 2012
@@ -24,6 +24,8 @@
<target name="download">
<!-- ensure the delition of the no longer used Danish tokenizer -->
<delete dir="${target.directory}" includes="da-token.bin" />
+ <!-- switched from perceptron to maxent for POS tagging for openNLP 1.5.2
-->
+ <delete file="${target.directory}/da-pos-perceptron.bin"
quiet="true"></delete>
<copy todir="${target.directory}" flatten="true">
<resources>
<!--
@@ -32,8 +34,9 @@
<url url="${model.url}/da-token.bin"/>
-->
<url url="${model.url}/da-sent.bin"/>
- <url url="${model.url}/da-pos-perceptron.bin"/>
- <!-- no Chunker for german
+ <!-- <url url="${model.url}/da-pos-perceptron.bin"/> -->
+ <url url="${model.url}/da-pos-maxent.bin"/>
+ <!-- no Chunker for da
<url url="${model.url}/da-chunker.bin"/>
-->
</resources>
Propchange:
incubator/stanbol/trunk/data/opennlp/lang/da/src/main/resources/org/apache/stanbol/data/opennlp/
------------------------------------------------------------------------------
--- svn:ignore (original)
+++ svn:ignore Thu May 24 14:27:40 2012
@@ -1,5 +1,3 @@
-da-token.bin
+da-pos-maxent.bin
da-sent.bin
-
-da-pos-perceptron.bin
Modified: incubator/stanbol/trunk/data/opennlp/lang/de/download_models.xml
URL:
http://svn.apache.org/viewvc/incubator/stanbol/trunk/data/opennlp/lang/de/download_models.xml?rev=1342285&r1=1342284&r2=1342285&view=diff
==============================================================================
--- incubator/stanbol/trunk/data/opennlp/lang/de/download_models.xml (original)
+++ incubator/stanbol/trunk/data/opennlp/lang/de/download_models.xml Thu May 24
14:27:40 2012
@@ -24,6 +24,8 @@
<!-- Removed the German Tokenizer, because it does sometime create "wrong"
tokens. In the absence of this model the SimpleTokenizer will be used by
default. -->
+ <!-- switched from perceptron to maxent for POS tagging for openNLP 1.5.2 -->
+ <delete file="${target.directory}/de-pos-perceptron.bin"
quiet="true"></delete>
<target name="download">
<!-- Added this delete to ensure that already downloaded versions of the
German Tokenizer models are removed. -->
@@ -33,7 +35,8 @@
<!-- url url="${model.url}/de-token.bin"/>
-->
<url url="${model.url}/de-sent.bin"/>
- <url url="${model.url}/de-pos-perceptron.bin"/>
+ <!-- url url="${model.url}/de-pos-perceptron.bin"/ -->
+ <url url="${model.url}/de-pos-maxent.bin"/>
<!-- no Chunker for german
<url url="${model.url}/de-chunker.bin"/>
-->
Propchange:
incubator/stanbol/trunk/data/opennlp/lang/de/src/main/resources/org/apache/stanbol/data/opennlp/
------------------------------------------------------------------------------
--- svn:ignore (original)
+++ svn:ignore Thu May 24 14:27:40 2012
@@ -1,5 +1,3 @@
de-sent.bin
-de-pos-perceptron.bin
-
-de-token.bin
+de-pos-maxent.bin
Modified: incubator/stanbol/trunk/data/opennlp/lang/en/download_models.xml
URL:
http://svn.apache.org/viewvc/incubator/stanbol/trunk/data/opennlp/lang/en/download_models.xml?rev=1342285&r1=1342284&r2=1342285&view=diff
==============================================================================
--- incubator/stanbol/trunk/data/opennlp/lang/en/download_models.xml (original)
+++ incubator/stanbol/trunk/data/opennlp/lang/en/download_models.xml Thu May 24
14:27:40 2012
@@ -22,11 +22,14 @@
</description>
<target name="download">
+ <!-- switched from perceptron to maxent for POS tagging for openNLP 1.5.2
-->
+ <delete file="${target.directory}/en-pos-perceptron.bin"
quiet="true"></delete>
<copy todir="${target.directory}" flatten="true">
<resources>
<url url="${model.url}/en-token.bin"/>
<url url="${model.url}/en-sent.bin"/>
- <url url="${model.url}/en-pos-perceptron.bin"/>
+ <!--url url="${model.url}/en-pos-perceptron.bin"/-->
+ <url url="${model.url}/en-pos-maxent.bin"/>
<url url="${model.url}/en-chunker.bin"/>
</resources>
</copy>
Modified: incubator/stanbol/trunk/data/opennlp/lang/nl/download_models.xml
URL:
http://svn.apache.org/viewvc/incubator/stanbol/trunk/data/opennlp/lang/nl/download_models.xml?rev=1342285&r1=1342284&r2=1342285&view=diff
==============================================================================
--- incubator/stanbol/trunk/data/opennlp/lang/nl/download_models.xml (original)
+++ incubator/stanbol/trunk/data/opennlp/lang/nl/download_models.xml Thu May 24
14:27:40 2012
@@ -22,11 +22,14 @@
</description>
<target name="download">
+ <!-- switched from perceptron to maxent for POS tagging for openNLP 1.5.2
-->
+ <delete file="${target.directory}/nl-pos-perceptron.bin"
quiet="true"></delete>
<copy todir="${target.directory}" flatten="true">
<resources>
<url url="${model.url}/nl-token.bin"/>
<url url="${model.url}/nl-sent.bin"/>
- <url url="${model.url}/nl-pos-perceptron.bin"/>
+ <!-- url url="${model.url}/nl-pos-perceptron.bin"/ -->
+ <url url="${model.url}/nl-pos-maxent.bin"/>
<!-- no Chunker for nl
<url url="${model.url}/nl-chunker.bin"/>
-->
Propchange:
incubator/stanbol/trunk/data/opennlp/lang/nl/src/main/resources/org/apache/stanbol/data/opennlp/
------------------------------------------------------------------------------
--- svn:ignore (original)
+++ svn:ignore Thu May 24 14:27:40 2012
@@ -1,5 +1,5 @@
nl-token.bin
-nl-pos-perceptron.bin
+nl-pos-maxent.bin
nl-sent.bin
Modified: incubator/stanbol/trunk/data/opennlp/lang/pt/download_models.xml
URL:
http://svn.apache.org/viewvc/incubator/stanbol/trunk/data/opennlp/lang/pt/download_models.xml?rev=1342285&r1=1342284&r2=1342285&view=diff
==============================================================================
--- incubator/stanbol/trunk/data/opennlp/lang/pt/download_models.xml (original)
+++ incubator/stanbol/trunk/data/opennlp/lang/pt/download_models.xml Thu May 24
14:27:40 2012
@@ -22,11 +22,14 @@
</description>
<target name="download">
+ <!-- switched from perceptron to maxent for POS tagging for openNLP 1.5.2
-->
+ <delete file="${target.directory}/pt-pos-perceptron.bin"
quiet="true"></delete>
<copy todir="${target.directory}" flatten="true">
<resources>
<url url="${model.url}/pt-token.bin"/>
<url url="${model.url}/pt-sent.bin"/>
- <url url="${model.url}/pt-pos-perceptron.bin"/>
+ <!-- url url="${model.url}/pt-pos-perceptron.bin"/ -->
+ <url url="${model.url}/pt-pos-maxent.bin"/>
<!-- no Chunker for pt
<url url="${model.url}/pt-chunker.bin"/>
-->
Propchange:
incubator/stanbol/trunk/data/opennlp/lang/pt/src/main/resources/org/apache/stanbol/data/opennlp/
------------------------------------------------------------------------------
--- svn:ignore (original)
+++ svn:ignore Thu May 24 14:27:40 2012
@@ -1,5 +1,5 @@
-pt-token.bin
+pt-pos-maxent.bin
-pt-pos-perceptron.bin
+pt-token.bin
pt-sent.bin
Modified: incubator/stanbol/trunk/data/opennlp/lang/sv/download_models.xml
URL:
http://svn.apache.org/viewvc/incubator/stanbol/trunk/data/opennlp/lang/sv/download_models.xml?rev=1342285&r1=1342284&r2=1342285&view=diff
==============================================================================
--- incubator/stanbol/trunk/data/opennlp/lang/sv/download_models.xml (original)
+++ incubator/stanbol/trunk/data/opennlp/lang/sv/download_models.xml Thu May 24
14:27:40 2012
@@ -26,6 +26,8 @@
<mapper type="merge" />
<!-- ensure the delition of the no longer used Swedish tokenizer -->
<delete dir="${target.directory}" includes="sv-token.bin" />
+ <!-- switched from perceptron to maxent for POS tagging for openNLP 1.5.2 -->
+ <delete file="${target.directory}/sv-pos-perceptron.bin"
quiet="true"></delete>
<target name="download">
<!-- Based on some testing the decision was to use the SimpleTokenizer for
@@ -43,11 +45,17 @@
</resources>
<mergemapper to="sv-sent.bin"/>
</copy>
- <copy toDir="${target.directory}/">
+ <!-- copy toDir="${target.directory}/">
<resources>
<url url="${model.url}/se-pos-perceptron.bin"/>
</resources>
<mergemapper to="sv-pos-perceptron.bin"/>
+ </copy -->
+ <copy toDir="${target.directory}/">
+ <resources>
+ <url url="${model.url}/se-pos-maxent.bin"/>
+ </resources>
+ <mergemapper to="sv-pos-maxent.bin"/>
</copy>
</target>
</project>
\ No newline at end of file
Propchange:
incubator/stanbol/trunk/data/opennlp/lang/sv/src/main/resources/org/apache/stanbol/data/opennlp/
------------------------------------------------------------------------------
--- svn:ignore (original)
+++ svn:ignore Thu May 24 14:27:40 2012
@@ -1,5 +1,3 @@
-sv-pos-perceptron.bin
-
sv-sent.bin
-sv-token.bin
+sv-pos-maxent.bin