Github user jkbradley commented on the pull request:
https://github.com/apache/spark/pull/5467#issuecomment-94348677
@MechCoder There are 2 pieces of code which we're talking about: (1)
converting the Map to a Matrix and (2) converting the Matrix to a Map. I
suppose that either approach would need both (assuming we supported the
constructor taking a Map in either approach).
I'm OK with leaving it as a to-do. I think the performance optimization
will be worthwhile at some point since I've heard of people complaining about
Word2Vec models getting too large to fit into memory. Keeping everything as a
Matrix without an intermediate Map representation would save some space for
large vocabs (and a bunch of object creation).
Would you prefer to leave it as a to-do item?
---
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]
