You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "ASF GitHub Bot (JIRA)" <ji...@apache.org> on 2018/09/27 09:26:00 UTC
[jira] [Commented] (IGNITE-9706) [ML] Update ignite-tensorflow to
support TensorFlow standalone client mode
[ https://issues.apache.org/jira/browse/IGNITE-9706?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16630041#comment-16630041 ]
ASF GitHub Bot commented on IGNITE-9706:
----------------------------------------
GitHub user dmitrievanthony opened a pull request:
https://github.com/apache/ignite/pull/4847
IGNITE-9706: Update ignite-tensorflow to support TensorFlow standalone client mode
TF_CONFIG variable, make user script to use TF_CLUSTER variable.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/gridgain/apache-ignite ignite-9706
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/ignite/pull/4847.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #4847
----
commit df6ee625ad571e3cdcec38bec2314fa09b2f2fa3
Author: Anton Dmitriev <dm...@...>
Date: 2018-09-26T15:13:53Z
IGNITE-9706: Make workers use run_standard_tensorflow_server and
TF_CONFIG variable, make user script to use TF_CLUSTER variable.
----
> [ML] Update ignite-tensorflow to support TensorFlow standalone client mode
> --------------------------------------------------------------------------
>
> Key: IGNITE-9706
> URL: https://issues.apache.org/jira/browse/IGNITE-9706
> Project: Ignite
> Issue Type: Improvement
> Components: ml
> Affects Versions: 2.7
> Reporter: Anton Dmitriev
> Assignee: Anton Dmitriev
> Priority: Blocker
> Fix For: 2.7
>
>
> We need to update "ignite-tensorflow" module to support TensorFlow standalone client mode described in in correspondent [documentation|https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/distribute#standalone-client-mode]. To make it we need to run TensorFlow standard servers on workers and avoid using "TF_CONFIG" variable in user script (use "TF_CLUSTER" instead).
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)