lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Cao Manh Dat (JIRA)" <>
Subject [jira] [Commented] (LUCENE-7274) Add LogisticRegressionDocumentClassifier
Date Thu, 02 Feb 2017 08:11:51 GMT


Cao Manh Dat commented on LUCENE-7274:

[~teofili] After review the patch, I'm afraid that we should close this issue as won't fix.

Because all classifiers in classification module are lazy learning methods and relied on Lucene
to quickly classify documents. They don't have any pre-trained model. Logistic Regression
in other way is eager learning method, so It need a pre-trained model to classify documents.
But the patch did not provide an api to train a logistic regression model, so it will be hard
for users to use {{LogisticRegressionDocumentClassifier}}.

BTW SOLR-8492 and SOLR-9252 provide an api for training a model. The trained model will be
stored as a document in Lucene index. So it will make Lucene depend on how Solr construct
that model, but I don't think it will be a good idea. 

So I think we can close this issue and create another issue like "an unify api for eager learning
method in classification module"

> Add LogisticRegressionDocumentClassifier
> ----------------------------------------
>                 Key: LUCENE-7274
>                 URL:
>             Project: Lucene - Core
>          Issue Type: Improvement
>          Components: modules/classification
>            Reporter: Cao Manh Dat
>            Assignee: Tommaso Teofili
>         Attachments: LUCENE-7274.patch
> Add LogisticRegressionDocumentClassifier for Lucene.

This message was sent by Atlassian JIRA

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message