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

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

yuxihu commented on a change in pull request #13980: [WIP] Use CPUPinned context in ImageRecordIOParser2
URL: https://github.com/apache/incubator-mxnet/pull/13980#discussion_r250755685
 
 

 ##########
 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:
   @ptrendx agree and thanks for your suggestion. How about changing to "The device id used to create CPUPinned context for internal NDArray."?

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