Github user avulanov commented on the pull request:
https://github.com/apache/spark/pull/1290#issuecomment-70717896
@tolgap Mean squared error from `ANNSuite` is not an error rate. In mnist
case, I guess you are looking for classification error, or (1 - accuracy). You
need to convert the output of the network to labels/classes and compare them to
the actual labels/classes. Number of misses divided by the total number of
samples will be the classification error. Alternatively, you might want to use
`ANNClassifier` and `MulticlassMetrics`, for example:
https://github.com/apache/spark/pull/1155#issuecomment-70714725
---
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
