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

[tf.data] Add documentation for supporting tf.data.experimental.AUTOTUNE as a value for the num_parallel_calls argument of map(), interleave(), and map_and_batch().

Browse Source 
PiperOrigin-RevId: 224381264
This commit is contained in:
Jiri Simsa 2018-12-06 12:00:29 -08:00 committed by TensorFlower Gardener
parent 0fb46cf913
commit 2947a28510
2 changed files with 10 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
tensorflow/python/data/experimental/ops/batching.py
View File

@ -626,9 +626,10 @@ def map_and_batch(map_func,
whether the last batch should be dropped in case its size is smaller than
desired; the default behavior is not to drop the smaller batch.
num_parallel_calls: (Optional.) A `tf.int32` scalar `tf.Tensor`,
representing the number of elements to process in parallel. If not
specified, `batch_size * num_parallel_batches` elements will be
processed in parallel.
representing the number of elements to process in parallel. If not
specified, `batch_size * num_parallel_batches` elements will be processed
in parallel. If the value `tf.data.experimental.AUTOTUNE` is used, then
the number of parallel calls is set dynamically based on available CPU.
Returns:
A `Dataset` transformation function, which can be passed to

8
tensorflow/python/data/ops/dataset_ops.py
View File

@ -947,7 +947,9 @@ class DatasetV2(object):
`self.output_types`) to another nested structure of tensors.
num_parallel_calls: (Optional.) A `tf.int32` scalar `tf.Tensor`,
representing the number elements to process in parallel. If not
specified, elements will be processed sequentially.
specified, elements will be processed sequentially. If the value
`tf.data.experimental.AUTOTUNE` is used, then the number of parallel
calls is set dynamically based on available CPU.
Returns:
Dataset: A `Dataset`.
@ -1058,7 +1060,9 @@ class DatasetV2(object):
num_parallel_calls: (Optional.) If specified, the implementation creates
a threadpool, which is used to fetch inputs from cycle elements
asynchronously and in parallel. The default behavior is to fetch inputs
from cycle elements synchronously with no parallelism.
from cycle elements synchronously with no parallelism. If the value
`tf.data.experimental.AUTOTUNE` is used, then the number of parallel
calls is set dynamically based on available CPU.
Returns:
Dataset: A `Dataset`.