[
https://issues.apache.org/jira/browse/SPARK-4036?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14362295#comment-14362295
]
Kai Sasaki commented on SPARK-4036:
-----------------------------------
[~mengxr] I'm thinking about design of CRF. I have a question. Current gradient
descent which is implemented in MLLib should be used in CRF, but current
{{Optimizer}} can receive only {{RDD\[Double, Vector\]}}. General CRF should
receive various type of labels and optimize it. Is there any plan to expand
{{Optimizer}} can optimize non-double labels(such as string or other). Or do
you have any other idea to train non-double labels in current {{Optimizer}}.
Thank you.
> Add Conditional Random Fields (CRF) algorithm to Spark MLlib
> ------------------------------------------------------------
>
> Key: SPARK-4036
> URL: https://issues.apache.org/jira/browse/SPARK-4036
> Project: Spark
> Issue Type: New Feature
> Components: MLlib
> Reporter: Guoqiang Li
> Assignee: Kai Sasaki
>
> Conditional random fields (CRFs) are a class of statistical modelling method
> often applied in pattern recognition and machine learning, where they are
> used for structured prediction.
> The paper:
> http://www.seas.upenn.edu/~strctlrn/bib/PDF/crf.pdf
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
