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

[TF FE] Update docs for TF FE (#17453)

Browse Source
This commit is contained in:
Roman Kazantsev
2023-05-10 14:30:23 +03:00
committed by GitHub
parent c378a8e912
commit 014eafda00
1 changed files with 5 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
docs/resources/tensorflow_frontend.md
View File

@@ -2,16 +2,14 @@
@sphinxdirective
TensorFlow Frontend is C++ based Frontend for conversion of TensorFlow models and is available as a preview feature starting from 2022.3.
That means that you can start experimenting with ``--use_new_frontend`` option passed to Model Optimizer to enjoy improved conversion time for limited scope of models
or directly loading TensorFlow models through ``read_model()`` method.
TensorFlow Frontend is a C++ based Frontend for converting TensorFlow models. MO now uses the TensorFlow Frontend as the default path for conversion to IR.
Also, the frontend allows loading TensorFlow models in SavedModel, MetaGraph, and frozen Protobuf formats directly via the ``read_model()`` method.
The current limitations:
* IRs generated by new TensorFlow Frontend are compatible only with OpenVINO API 2.0
* There is no full parity yet between legacy Model Optimizer TensorFlow Frontend and new TensorFlow Frontend so primary path for model conversion is still legacy frontend
* Model coverage and performance is continuously improving so some conversion phase failures, performance and accuracy issues might occur in case model is not yet covered.
Known unsupported models: object detection models and all models with transformation configs, models with TF1/TF2 control flow, Complex type and training parts
* ``read_model()`` method supports only ``*.pb`` format while Model Optimizer (or ``convert_model`` call) will accept other formats as well which are accepted by existing legacy frontend
* There is no full parity between the legacy frontend and the new frontend in MO. Known limitations compared to the legacy approach are:
TF1 Control flow, Complex types, models requiring config files and old python extensions. The solution detects unsupported functionalities and provides fallback.
To force the use of the legacy frontend, ``--use_legacy_frontend`` must be specified.
@endsphinxdirective