Github user jkbradley commented on the pull request:
https://github.com/apache/spark/pull/5467#issuecomment-94582162
It just occurred to me that we're converting from Float to Double. I'm not
sure historically why Word2Vec used Float, but I'm worrying now about switching
since it will double model sizes. (I'm sorry I didn't think about this
earlier!)
This PR should still be doable, but you would need to store an Array[Float]
instead of the Matrix type. You would also need to use
```mllib.linalg.BLAS.nativeBLAS``` to make the BLAS calls.
What do you think?
---
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]
