projects / platform / upstream / tensorflow.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: c5a16fa) | patch
[TPUClusterResolver] Start a TFServer when running in GKE
authorBrennan Saeta <saeta@google.com>
Fri, 6 Apr 2018 16:26:08 +0000 (09:26 -0700)
committerTensorFlower Gardener <gardener@tensorflow.org>
Fri, 6 Apr 2018 16:28:43 +0000 (09:28 -0700)
commit218647db25d1e754ad85fd1fa8a0960b82ae83bf
tree2e43dd2461814c84cc0db55662188a803dada27btree | snapshot
parentc5a16fa1c91a0d1cf3d5b432d70b4e8fe47b88cdcommit | diff
[TPUClusterResolver] Start a TFServer when running in GKE

This change allows advanced input pipelines (e.g. StreamingFilesDataset, or split-pipelines that use py_func's) to run in GKE- and GKE-like enviornments.

PiperOrigin-RevId: 191897639
tensorflow/contrib/cluster_resolver/python/training/tpu_cluster_resolver.py diff | blob | history
tensorflow/contrib/cluster_resolver/python/training/tpu_cluster_resolver_test.py diff | blob | history
Domain: Machine Learning / ML Framework;