experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[tf.data] Fix documentation.

Browse Source 
PiperOrigin-RevId: 331623200
Change-Id: Ib7e4d5b050878d119924e428d11e53902fdf3cf3
This commit is contained in:
Jiri Simsa 2020-09-14 13:57:20 -07:00 committed by TensorFlower Gardener
parent 00f20cc833
commit 3ad38e1fea
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
tensorflow/python/data/ops/multi_device_iterator_ops.py
View File

@ -223,7 +223,7 @@ class MultiDeviceIterator(object):
dataset: The input dataset to be iterated over.
devices: The list of devices to fetch data to.
max_buffer_size: Maximum size of the host side per device buffer to keep.
prefetch_buffer_size: if > 1, then we setup a buffer on each device to
prefetch_buffer_size: if > 0, then we setup a buffer on each device to
prefetch into.
source_device: The host device to place the `dataset` on. In order to
prevent deadlocks, if the prefetch_buffer_size is greater than the
@ -481,7 +481,7 @@ class OwnedMultiDeviceIterator(composite_tensor.CompositeTensor):
dataset: The input dataset to be iterated over.
devices: The list of devices to fetch data to.
max_buffer_size: Maximum size of the host side per device buffer to keep.
prefetch_buffer_size: if > 1, then we setup a buffer on each device to
prefetch_buffer_size: if > 0, then we setup a buffer on each device to
prefetch into.
source_device: The host device to place the `dataset` on. In order to
prevent deadlocks, if the prefetch_buffer_size is greater than the