[FLINK-24808] Support IterationListener for per-round operators

• Type: Sub-task
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 0.1.0
• Fix Version/s: ml-2.0.0
• Component/s: Library / Machine Learning
• Labels: pull-request-available
• Assignee: Yun Gao
• Reporter: Yun Gao
• Created: 06/Nov/21 04:10
• Updated: 15/Dec/21 03:11
• Resolved: 15/Nov/21 03:26

Description

We would need to also support IterationListener for the per-round operators. The difference between onEpochWatermarkIncremented and endInput for per-round operators are if it would be called at the last round (epoch = Integer.MAX_VALUE). This is necessary for some per-round operators that outputs final result.

Issue Links

• links to: GitHub Pull Request #26