[
https://issues.apache.org/jira/browse/OPENNLP-1304?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17123678#comment-17123678
]
Jeffrey T. Zemerick commented on OPENNLP-1304:
----------------------------------------------
[~markus17] Can you let me know a rough size of how much training data was used
to create the model?
> NullPointerException in LemmatizerME
> ------------------------------------
>
> Key: OPENNLP-1304
> URL: https://issues.apache.org/jira/browse/OPENNLP-1304
> Project: OpenNLP
> Issue Type: Bug
> Components: Lemmatizer
> Affects Versions: 1.9.2
> Reporter: Markus Jelsma
> Priority: Major
> Attachments: ja-lemma.bin
>
>
> In our unit tests we have a basic lemmatizer model for Japanese. The
> Lemmatizer throws:
> {code:java}
> java.lang.NullPointerException
> at
> opennlp.tools.lemmatizer.LemmatizerME.predictSES(LemmatizerME.java:116)
> at
> opennlp.tools.lemmatizer.LemmatizerME.lemmatize(LemmatizerME.java:91) {code}
> for the following input:
> {code:java}
> tokens: デン, ・, ヘルデル
> tags: PROPN, SYM, PROPN {code}
> I attached the used lemmatizer model in case if needed.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
