You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@opennlp.apache.org by "Rodrigo Agerri (JIRA)" <ji...@apache.org> on 2016/12/22 16:09:58 UTC

[jira] [Commented] (OPENNLP-900) DictionaryLemmatizer dictionary and LemmatizerME training format different

    [ https://issues.apache.org/jira/browse/OPENNLP-900?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15770400#comment-15770400 ] 

Rodrigo Agerri commented on OPENNLP-900:
----------------------------------------

Sure, I do not see any problem. The dictionary lemmatizer had that order because of historical reasons (most existing dictionaries came in that format).

> DictionaryLemmatizer dictionary and LemmatizerME training format different
> --------------------------------------------------------------------------
>
>                 Key: OPENNLP-900
>                 URL: https://issues.apache.org/jira/browse/OPENNLP-900
>             Project: OpenNLP
>          Issue Type: Bug
>          Components: Lemmatizer
>    Affects Versions: 1.6.0, 1.7.0
>            Reporter: Daniel Russ
>            Priority: Minor
>             Fix For: 1.7.0
>
>         Attachments: LemmaTest1.java
>
>
> The LemmatizerME training data has a format of word\tpos\tlemma.
> The DictionaryLemmatizer format is word\tlemma\tpos.
> Can we make them the same.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)