IntenseWebs/openvino
3
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[DOC] Update list of TF formats imported from memory. (#20834)

Browse Source 
* Update list of TF formats.

* Minor correction.

* Added comment.

* Update docs/articles_en/openvino_workflow/model_preparation/Convert_Model_From_TensorFlow.md

Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>

* Model changed.

* Update docs/articles_en/openvino_workflow/model_preparation/Convert_Model_From_TensorFlow.md

Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>

---------

Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>
This commit is contained in:
Anastasiia Pnevskaia 2023-11-02 14:31:08 +01:00 committed by GitHub
parent f8cd53bb77
commit 3f5f923a70
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 27 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

45
docs/articles_en/openvino_workflow/model_preparation/Convert_Model_From_TensorFlow.md
View File

@ -196,6 +196,31 @@ Converting TensorFlow Models from Memory Using Python API
Model conversion API supports passing TensorFlow/TensorFlow2 models directly from memory.
* ``Trackable``. The object returned by ``hub.load()`` can be converted to ``ov.Model`` with ``convert_model()``.
.. code-block:: py
:force:
import tensorflow_hub as hub
import openvino as ov
model = hub.load("https://tfhub.dev/google/movenet/singlepose/lightning/4")
ov_model = ov.convert_model(model)
* ``tf.function``
.. code-block:: py
:force:
@tf.function(
input_signature=[tf.TensorSpec(shape=[1, 2, 3], dtype=tf.float32),
tf.TensorSpec(shape=[1, 2, 3], dtype=tf.float32)])
def func(x, y):
return tf.nn.sigmoid(tf.nn.relu(x + y))
import openvino as ov
ov_model = ov.convert_model(func)
* ``tf.keras.Model``
.. code-block:: py
@ -205,7 +230,7 @@ Model conversion API supports passing TensorFlow/TensorFlow2 models directly fro
model = tf.keras.applications.ResNet50(weights="imagenet")
ov_model = ov.convert_model(model)
* ``tf.keras.layers.Layer``. Requires saving model to TensorFlow ``saved_model`` file format and then loading to ``openvino.convert_model``. Saving to the file and then restoring is required due to a known bug in ``openvino.convert_model`` that ignores model signature.
* ``tf.keras.layers.Layer``. The ``ov.Model`` converted from ``tf.keras.layers.Layer`` does not contain original input and output names. So it is recommended to convert the model to ``tf.keras.Model`` before conversion or use ``hub.load()`` for TensorFlow Hub models.
.. code-block:: py
:force:
@ -214,10 +239,8 @@ Model conversion API supports passing TensorFlow/TensorFlow2 models directly fro
import openvino as ov
model = hub.KerasLayer("https://tfhub.dev/google/imagenet/mobilenet_v1_100_224/classification/5")
model.build([None, 224, 224, 3])
model.save('mobilenet_v1_100_224') # use a temporary directory
ov_model = ov.convert_model(model)
ov_model = ov.convert_model('mobilenet_v1_100_224')
* ``tf.Module``. Requires setting shapes in ``input`` parameter.
@ -270,20 +293,6 @@ Model conversion API supports passing TensorFlow/TensorFlow2 models directly fro
import openvino as ov
ov_model = ov.convert_model(model)
* ``tf.function``
.. code-block:: py
:force:
@tf.function(
input_signature=[tf.TensorSpec(shape=[1, 2, 3], dtype=tf.float32),
tf.TensorSpec(shape=[1, 2, 3], dtype=tf.float32)])
def func(x, y):
return tf.nn.sigmoid(tf.nn.relu(x + y))
import openvino as ov
ov_model = ov.convert_model(func)
* ``tf.compat.v1.session``
.. code-block:: py