NullPointerException

[Nulik]

Hi,

I downloaded FactorIE yesterday, just getting into it. Tried a few examples, and when execution "classify" test I get null pointer exception.
Reproduce it like this, with the 20 newsgroups data test:

[niko@dev1 factorie]$ ./bin/fac classify --read-text-dirs ../20_newsgroups/sci.space/ --write-classifier mymodel.factorie
java -Xmx6g -ea -Djava.awt.headless=true -Dfile.encoding=UTF-8 -server -classpath ./src/main/resources:./target/classes:./target/factorie_2.11-1.2-SNAPSHOT-nlp-jar-with-dependencies.jar
Exception in thread "main" java.lang.NullPointerException
    at scala.collection.mutable.ArrayOps$ofRef$.length$extension(ArrayOps.scala:192)
    at scala.collection.mutable.ArrayOps$ofRef.length(ArrayOps.scala:192)
    at scala.collection.IndexedSeqOptimized$class.foreach(IndexedSeqOptimized.scala:32)
    at scala.collection.mutable.ArrayOps$ofRef.foreach(ArrayOps.scala:186)
    at scala.collection.TraversableLike$WithFilter.foreach(TraversableLike.scala:777)
    at cc.factorie.app.classify.Classify$$anonfun$main$4.apply(Classify.scala:231)
    at cc.factorie.app.classify.Classify$$anonfun$main$4.apply(Classify.scala:228)
    at scala.collection.IndexedSeqOptimized$class.foreach(IndexedSeqOptimized.scala:33)
    at scala.collection.mutable.ArrayOps$ofRef.foreach(ArrayOps.scala:186)
    at cc.factorie.app.classify.Classify$.main(Classify.scala:228)
    at cc.factorie.app.classify.Classify.main(Classify.scala)
[niko@dev1 factorie]$

The dataset can be downloaded from this page:
http://www.cs.cmu.edu/~mccallum/bow/rainbow/

[Emma Strubell]

I believe --read-text-dirs expects either a single directory containing subdirectories representing classes, or a comma-separated list of the directories representing classes. Try just passing "../20_newsgroups".

Emma

--
--
Factorie Discuss group.
To post, email: dis...@factorie.cs.umass.edu
To unsubscribe, email: discuss+u...@factorie.cs.umass.edu

[Nulik]

On Thursday, February 18, 2016 at 9:39:25 AM UTC-6, Emma Strubell wrote:

I believe --read-text-dirs expects either a single directory containing subdirectories representing classes, or a comma-separated list of the directories representing classes. Try just passing "../20_newsgroups".

thanks. I removed the slash, but now I get new error: MalformedInputException: Input length = 1

[niko@dev1 factorie]$ ./bin/fac classify --read-text-dirs ../20_newsgroups --write-classifier mymodel.factorie

java -Xmx6g -ea -Djava.awt.headless=true -Dfile.encoding=UTF-8 -server -classpath ./src/main/resources:./target/classes:./target/factorie_2.11-1.2-SNAPSHOT-nlp-jar-with-dependencies.jar

Exception in thread "main" java.nio.charset.MalformedInputException: Input length = 1
    at java.nio.charset.CoderResult.throwException(CoderResult.java:281)
    at sun.nio.cs.StreamDecoder.implRead(StreamDecoder.java:339)
    at sun.nio.cs.StreamDecoder.read(StreamDecoder.java:178)
    at java.io.InputStreamReader.read(InputStreamReader.java:184)
    at java.io.BufferedReader.read1(BufferedReader.java:210)
    at java.io.BufferedReader.read(BufferedReader.java:286)
    at java.io.Reader.read(Reader.java:140)
    at scala.io.BufferedSource.mkString(BufferedSource.scala:96)
    at cc.factorie.app.classify.Classify$.cc$factorie$app$classify$Classify$$fileToString$1(Classify.scala:150)
    at cc.factorie.app.classify.Classify$$anonfun$main$4$$anonfun$apply$2.apply(Classify.scala:239)
    at cc.factorie.app.classify.Classify$$anonfun$main$4$$anonfun$apply$2.apply(Classify.scala:231)
    at scala.collection.TraversableLike$WithFilter$$anonfun$foreach$1.apply(TraversableLike.scala:778)

[Emma Strubell]

That sounds like an encoding issue. Are you sure the files are actually encoded as UTF-8?

--

[Nulik]

On Thursday, February 18, 2016 at 11:39:27 AM UTC-6, Emma Strubell wrote:

That sounds like an encoding issue. Are you sure the files are actually encoded as UTF-8?

no, the 20_newsgroups data set is ASCII. I encoded and it worked. Thanks. 

Therefore, everything has to be encoded ? Because I ran the topics on 20_newsgroups and it didn't complain about utf8.