mo.utils.error.Error: Unexpected exception happened during extracting attributes for node i1

[vpotnis]

Hi,  I'm trying to generate a OpenVINO model using TensorFlow Model (.pb) as an input.  Please see the attached files for reference.

I get the following error while running model optimizer script (which I don't get with another model that expects int inputs and generates int output, note that this one expects uint8 as input)

[ ERROR ]  Unexpected exception happened during extracting attributes for node i1.
Original exception message: uint8_val
)
  File "/opt/intel/openvino_2020.1.023/deployment_tools/model_optimizer/mo/front/tf/common.py", line 26, in <lambda>
    ('DT_UINT8', np.uint8, lambda pb: pb.uint8_val, lambda x: np.uint8(x)),
AttributeError: uint8_val

Here's how I'm running the mo_tf script:
$ /opt/intel/openvino/deployment_tools/model_optimizer$ python3.6 mo_tf.py --input_model=test_2d_byte_to_2d_byte.pb --input_shape=[3,2] --output_dir=/tmp/test --log=DEBUG

The complete run log is attached as err.log.txt. The tar ball test.tar.gz contains the required tensorflow test pb file.

[Iffa_Intel]

Greetings,

Fyi,the model that you are trying to use right now is currently unsupported.

Instead, perhaps you can try these:

https://docs.openvinotoolkit.org/latest/openvino_docs_MO_DG_prepare_model_convert_model_Convert_Model_From_TensorFlow.html

Sincerely,

Iffa

[Iffa_Intel]

Greetings,

Intel will no longer monitor this thread since we have provided a solution. If you need any additional information from Intel, please submit a new question.

Sincerely,

Iffa