[GitHub] ptrendx commented on a change in pull request #13980: [WIP] Use CPUPinned context in ImageRecordIOParser2

[GitBox <gi...@apache.org>]

ptrendx commented on a change in pull request #13980: [WIP] Use CPUPinned context in ImageRecordIOParser2
URL: https://github.com/apache/incubator-mxnet/pull/13980#discussion_r250695875
 
 

 ##########
 File path: src/io/image_iter_common.h
 ##########
 @@ -163,6 +165,8 @@ struct ImageRecParserParam : public dmlc::Parameter<ImageRecParserParam> {
         .describe("Virtually partition the data into these many parts.");
     DMLC_DECLARE_FIELD(part_index).set_default(0)
         .describe("The *i*-th virtual partition to be read.");
+    DMLC_DECLARE_FIELD(device_id).set_default(0)
+        .describe("The device id on which to store the data.");
 
 Review comment:
   Could you expand this documentation text? It is not clear what kind of "device" this refers to and why would you want to set it. Especially since the reader produces data for all of the devices, user will probably be confused which one to choose.

----------------------------------------------------------------
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