Github user avulanov commented on the pull request:
https://github.com/apache/spark/pull/1290#issuecomment-103975077
@loachli looks good! If you make more iterations for configurations with
hidden layers, it will converge with accuracy about 0.95. I suggest using
batches (100), plugging native BLAS and persisting the training set to speed up
the computations. Also, with such size of dataset you might not take advantage
of several nodes that you have, because all data might reside on one node.
---
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]
