[ 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 <dmitrievanthony@...> 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)