You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Anton Dmitriev (JIRA)" <ji...@apache.org> on 2019/01/21 10:05:00 UTC
[jira] [Commented] (IGNITE-10996) [ML] Add additional host and port
into IgniteDataset to retrieve schema
[ https://issues.apache.org/jira/browse/IGNITE-10996?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16747807#comment-16747807 ]
Anton Dmitriev commented on IGNITE-10996:
-----------------------------------------
Please, have a look at the following PR in [tensorflow/io|https://github.com/tensorflow/io] repository: [https://github.com/tensorflow/io/pull/59|https://github.com/tensorflow/io/pull/59].
> [ML] Add additional host and port into IgniteDataset to retrieve schema
> -----------------------------------------------------------------------
>
> Key: IGNITE-10996
> URL: https://issues.apache.org/jira/browse/IGNITE-10996
> Project: Ignite
> Issue Type: Improvement
> Components: ml
> Affects Versions: 2.8
> Reporter: Anton Dmitriev
> Assignee: Anton Dmitriev
> Priority: Major
> Fix For: 2.8
>
>
> Currently [IgniteDataset|https://github.com/tensorflow/io/tree/master/tensorflow_io/ignite#distributed-in-memory-datasource] provides ability to specify only a single host/port for data schema extraction and for data loading. This limitation doesn't allow to use [Dataset.interleave(...)|https://www.tensorflow.org/api_docs/python/tf/data/Dataset#interleave] and [tf.data.experimental.parallel_interleave|https://www.tensorflow.org/api_docs/python/tf/data/experimental/parallel_interleave] function for parallel data loading.
> We need to add ability to specify host/port separately for schema extraction and to data loading.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)