[LUCENE-5338] Let classifiers filter unlabeled documents during training

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 4.5
• Fix Version/s: 4.6, 6.0
• Component/s: modules/classification
• Lucene Fields: New
• Assignee: Tommaso Teofili
• Reporter: Tommaso Teofili
• Created: 11/Nov/13 10:35
• Updated: 28/Aug/22 13:56
• Resolved: 11/Nov/13 13:27

Description

Only labeled (with 'class' field) documents should be used during training and therefore each classifier should filter (and not fail when run against) them.