Github user AnthonyTruchet commented on the issue:
https://github.com/apache/spark/pull/16038
THe big doubt I have is on *Of course, the first call to axpy should
produce a dense vector, but, that's already on the executor*: the other operand
is Sparse and has to be sparse, and this axpy call between to sparse vectors
has to return a sparse vector. And we do not want to cast at the end because
the performance penalties of using sparse instead of dense will already have
been paid.
Fully exploring the trade-off between sparse and dense aggregation os
ot-of-scope and likely very application specific. I'll have a new look at the
problem trying to manage a work around along the line of Sparse vectors, but
this will not solve the typing issue as far as i can see for now.
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
