Error converting SqueezeDet from TensorFlow pb to OpenVINO IR

[Gouveia__César]

Hi,

I tried to convert a Tensorflow model (pb) into an OpenVINO IR (bin, xml) however the following error appears:

[ ERROR ]  Cannot infer shapes or values for node "lambda_17/Select".
[ ERROR ]
[ ERROR ]
[ ERROR ]  It can happen due to bug in custom shape infer function <function Select.infer at 0x0000028738CE8840>.
[ ERROR ]  Or because the node inputs have incorrect values/shapes.
[ ERROR ]  Or because input shapes are incorrect (embedded to the model or passed via --input_shape).
[ ERROR ]  Run Model Optimizer with --log_level=DEBUG for more information.
[ ERROR ]

Bellow I attached the full logs (using the --log_level DEBUG flag) and I can share the model privately for you to reproduce and test the behavior, just send me a PM.

Thanks in advance,

César.

[Luis_at_Intel]

Hi Gouveia, César,

Please share your model via PM so we can test and reproduce the behavior. Also share any details you have about the model (topology, custom trained, etc.) and your environment config (versions of OpenVINO, OS, python, TF, etc.)

Regards,

Luis