Github user jkbradley commented on the pull request:
https://github.com/apache/spark/pull/7099#issuecomment-119734870
@feynmanliang Apart from those comments, the main other comment I have is
that we should force eager evaluation.
@dbtsai Per your discussion with @mengxr this is being refactored to force
eager evaluation for cheap metrics (ones requiring little memory and only 1
pass over the data) at the end of training. Once we add more expensive
metrics, we can add parameters to LinearRegression so that the user can specify
which of those expensive metrics to compute and save during training. We
decided it was not worth adding parameters for cheap metrics, but it will be
important for more expensive metrics.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
