Github user feynmanliang commented on the pull request:
https://github.com/apache/spark/pull/6795#issuecomment-118256745
Actually, I think the performance gains are becausethe default `HashingTF`
has 2^20 bins but your test data has at most 20000 unique words.
If you configure the `HashingTF` properly (e.g. `val tf = new
HashingTF(25000)`), is the performance difference still significant?
Also, are there other places besides an improperly configured `HashingTF`
where this may be used? Do you know of any real world use cases or papers using
this technique?
---
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]
