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

Deprecate tf.distribute.Strategy.experimental_make_numpy_iterator

Browse Source 
before it is in an official TensorFlow release.

PiperOrigin-RevId: 239525447
This commit is contained in:
A. Unique TensorFlower 2019-03-20 19:51:54 -07:00 committed by TensorFlower Gardener
parent c261779aff
commit dca741b737
1 changed files with 3 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
tensorflow/python/distribute/distribute_lib.py
View File

@ -425,10 +425,13 @@ class DistributionStrategy(object):
return self.extended._make_input_fn_iterator( # pylint: disable=protected-access
input_fn, replication_mode=replication_mode)
@doc_controls.do_not_generate_docs # DEPRECATED
def experimental_make_numpy_iterator(
self, numpy_input, batch_size, num_epochs=1, shuffle=1024, session=None):
"""Makes an iterator for input provided via a nest of numpy arrays.
DEPRECATED: Use `extended.experimental_make_numpy_dataset` instead.
Args:
numpy_input: A nest of NumPy input arrays that will be distributed evenly
across all replicas. Note that lists of Numpy arrays are stacked,