You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2018/03/29 18:51:25 UTC
[GitHub] piiswrong commented on a change in pull request #10285: [MXNET-241] Module API for distributed training w/ row_sparse weight
piiswrong commented on a change in pull request #10285: [MXNET-241] Module API for distributed training w/ row_sparse weight
URL: https://github.com/apache/incubator-mxnet/pull/10285#discussion_r178149942
##########
File path: python/mxnet/module/base_module.py
##########
@@ -217,6 +219,10 @@ def score(self, eval_data, eval_metric, num_batch=None, batch_end_callback=None,
epoch : int
Defaults to 0. For compatibility, this will be passed to callbacks (if any).
During training, this will correspond to the training epoch number.
+ row_id_generator : A callback function
+ The function takes `data_batch` as an input and returns a dict of
+ str -> NDArray. The resulting dict is used for pulling row_sparse
+ parameters from the kvstore.
Review comment:
needs more explanation. what does str represent? what does NDArray represent? why do we need to pull row_sparse parameters?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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