Github user debasish83 commented on the pull request:
https://github.com/apache/spark/pull/1290#issuecomment-83326047
@avulanov thanks for the reference...I am more interested in fully
connected layers first...Let's discuss it in few weeks...I will ping...
I am understanding LDA's graphx implementation...LDA has 2 partitions
(document and words) but I am considering 2 partitions which both have features
and we add 10K nodes more in this graph (hidden layer)...now the issue is that
every data point needs to be passed over this graph but assuming the data is
sparse (which is often the case for these models) can we do something ? If the
data is not sparse and has 3M columns then I don't think anything will work...
---
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]
