You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@madlib.apache.org by GitBox <gi...@apache.org> on 2019/05/01 19:19:16 UTC
[GitHub] [madlib] njayaram2 opened a new pull request #381: DL: Add input
preprocessor for validation data
njayaram2 opened a new pull request #381: DL: Add input preprocessor for validation data
URL: https://github.com/apache/madlib/pull/381
JIRA: MADLIB-1333
This PR makes the following changes:
1) Rename minibatch_preprocessor_dl to training_preprocessor_dl.
2) Code refactoring to move it from utilities module to deep_learning.
3) Add a new function named validation_preprocessor_dl that can be used
to preprocess validation data. The independent var in validation data is
minibatched and normalized while the dependent var is 1-hot encoded. We
do not distribute the minibatched data randomly, since this data is only
used for evaluation and not training.
4) Necessary changes to get existing tests to green.
5) Dev check tests.
6) User and online doc changes.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services