diff options
author | EFanZh <efanzh@gmail.com> | 2018-09-28 15:20:26 +0800 |
---|---|---|
committer | EFanZh <efanzh@gmail.com> | 2018-09-28 15:20:26 +0800 |
commit | d0690d46466bf0393ad65544d1e8c55e948df133 (patch) | |
tree | d0ccbd2dbf7c3752e991881a93684403bdd39f13 /tensorflow/contrib/distribute | |
parent | 6ebe9baae06c06d0a70a424a55c78f5af07b49f7 (diff) |
Fix some documentation errors
Diffstat (limited to 'tensorflow/contrib/distribute')
-rw-r--r-- | tensorflow/contrib/distribute/python/mirrored_strategy.py | 5 |
1 files changed, 3 insertions, 2 deletions
diff --git a/tensorflow/contrib/distribute/python/mirrored_strategy.py b/tensorflow/contrib/distribute/python/mirrored_strategy.py index 504f45a695..c0861da567 100644 --- a/tensorflow/contrib/distribute/python/mirrored_strategy.py +++ b/tensorflow/contrib/distribute/python/mirrored_strategy.py @@ -318,12 +318,13 @@ class MirroredStrategy(distribute_lib.DistributionStrategy): [TensorFlow's documentation](https://www.tensorflow.org/deploy/distributed). The distribution strategy inherits these concepts as well and in addition to that we also clarify several more concepts: - * **In-graph replication**: the `client` creates a single `tf.Graph` that + + * **In-graph replication**: the `client` creates a single `tf.Graph` that specifies tasks for devices on all workers. The `client` then creates a client session which will talk to the `master` service of a `worker`. Then the `master` will partition the graph and distribute the work to all participating workers. - * **Worker**: A `worker` is a TensorFlow `task` that usually maps to one + * **Worker**: A `worker` is a TensorFlow `task` that usually maps to one physical machine. We will have multiple `worker`s with different `task` index. They all do similar things except for one worker checkpointing model variables, writing summaries, etc. in addition to its ordinary work. |