ScatterND Shape Cannot be Infered

[CollinBrake]

Hi,

I am attempting to convert a custom YOLOv3 network from ONNX format to OpenVino IR format. When I import the model to the Deep Learning Workbench and run the conversion, the following error results (complete log attached):

Cannot infer shapes or values for node "ScatterND_1248".
index 16 is out of bounds for axis 0 with size 16
It can happen due to bug in custom shape infer function <function ScatterNDUpdate.infer at 0x7fe26c757c80>.
Or because the node inputs have incorrect values/shapes.
Or because input shapes are incorrect (embedded to the model or passed via --input_shape).

Is there a way to pass this input shape to the model optimizer in the Deep Learning Workbench GUI?

[Syamimi_Intel]

Hi COLLIN BRAKE,

Thank you for reaching out. This error is might be due to your input shapes value are incorrect. Maybe you can have a try to specify the input shape in the Deep Learning Workbench GUI by referring to this Import ONNX* MobileNet v2 Tutorial.

Regards,

Syamimi

[Syamimi_Intel]

Hi COLLIN BRAKE,

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

Regards,

Syamimi