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

Mark tf.compat.v1.lite.experimental.get_potentially_supported_ops deprecated.

Browse Source 
PiperOrigin-RevId: 360891012
Change-Id: I3e6ef6b2fa5d98e2a6828bde4af03e2e5225e605
This commit is contained in:
Tian Lin 2021-03-04 05:16:01 -08:00 committed by TensorFlower Gardener
parent fab8fb9b9a
commit a5629c8368
3 changed files with 12 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
RELEASE.md
View File

@ -108,8 +108,11 @@
ML authoring is generally discouraged. ML authoring is generally discouraged.
* Add support for static hash tables through * Add support for static hash tables through
`TFLiteConverter.from_saved_model`. `TFLiteConverter.from_saved_model`.
* Quantized x86 execution defaults to Ruy GEMM library for platforms with * Quantized x86 execution defaults to Ruy GEMM library for platforms with
AVX support. AVX support.
* Deprecate `tf.compat.v1.lite.experimental.get_potentially_supported_ops`.
Use `tf.lite.TFLiteConverter` directly to check whether a model is
convertible.
* TF Core: * TF Core:
* Corrected higher-order gradients of control flow constructs (`tf.cond`, * Corrected higher-order gradients of control flow constructs (`tf.cond`,

1
tensorflow/lite/experimental/tensorboard/BUILD
View File

@ -13,6 +13,7 @@ py_library(
deps = [ deps = [
"//tensorflow/lite/python:wrap_toco", "//tensorflow/lite/python:wrap_toco",
"//tensorflow/python:util", "//tensorflow/python:util",
"//tensorflow/python/util:tf_export",
], ],
) )

6
tensorflow/lite/experimental/tensorboard/ops_util.py
View File

@ -20,6 +20,7 @@ from __future__ import print_function
import collections import collections
from tensorflow.lite.python import wrap_toco from tensorflow.lite.python import wrap_toco
from tensorflow.python.util import deprecation
from tensorflow.python.util.tf_export import tf_export from tensorflow.python.util.tf_export import tf_export
@ -32,6 +33,11 @@ class SupportedOp(collections.namedtuple("SupportedOp", ["op"])):
@tf_export(v1=["lite.experimental.get_potentially_supported_ops"]) @tf_export(v1=["lite.experimental.get_potentially_supported_ops"])
@deprecation.deprecated(
None, "Deprecated in TF 2.4 and targeted to remove after TF 2.5. This"
"experimental function in TF v1 is to get a list of op names without real "
"conversion. To check whether a model can be convertable or not indeed, "
"please run `tf.lite.TFLiteConverter`.")
def get_potentially_supported_ops(): def get_potentially_supported_ops():
"""Returns operations potentially supported by TensorFlow Lite. """Returns operations potentially supported by TensorFlow Lite.