Github user viirya commented on the pull request:
https://github.com/apache/spark/pull/3643#issuecomment-67450454
Calculating 20000 squared distances between the vectors of 20000 dims:
* `DenseVector` vs. `SparseVector`
* breezeSquaredDistance: ~25 secs
* This PR: ~9.5 secs
* `DenseVector` vs. `DenseVector`
* breezeSquaredDistance: ~18.5 secs
* This PR: ~11 secs
* `SparseVector` vs. `SparseVector` (same indices length)
* breezeSquaredDistance: ~35.5 secs
* This PR: ~16.5 secs
* `SparseVector` vs. `SparseVector` (different indices length)
* breezeSquaredDistance: ~39.8 secs
* This PR: ~19.8 secs
The timing test results are not very accurate. But it shows the relative
improvement of this PR comparing to `breezeSquaredDistance`.
---
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]
