Supported TensorFlow 2 model format and version for OpenVINO™ Toolkit
-
Unable to convert non-frozen TensorFlow model version 2.4.1 with OpenVINO™ Toolkit 2020.3:
python3 mo_tf.py --input_model <INPUT_MODEL>.pb -
Received error:
[ ERROR ] Exception occurred during running replacer "REPLACEMENT_ID" (<class 'extensions.load.tf.loader.TFLoader'>): Unexpected exception happened during extracting attributes for node Adam/dense_11/bias/v/Read/ReadVariableOp.
Original exception message: 'ascii' codec can't decode byte 0xcc in position 1: ordinal not in range(128)
TensorFlow 2.x models were supported starting from OpenVINO™ Toolkit 2020.4.
Upgrade to OpenVINO™ Toolkit 2020.4 or later.
TensorFlow 2.x officially supports two model formats: SavedModel and Keras H5 (or HDF5)
Refer to Convert TensorFlow 2 to IR
Refer to Release Notes for Intel® Distribution of OpenVINO Toolkit 2020