Internal Error addScale condition: shift.count > 0 ? (shift.values != nullptr) : (shift.values == nullptr) )

While I am try to build Engine, I am encountering following error

[E] [TRT] 3: [network.cpp::addScale::616] Error Code 3: Internal Error (Parameter check failed at: optimizer/api/network.cpp::addScale::616, condition: shift.count > 0 ? (shift.values != nullptr) : (shift.values == nullptr)
)

Here is onnx file

Hi,

Which TensorRT version do you use?
We test your model with TensorRT v8.0 from JetPack 4.6.

[02/25/2022-00:00:24] [E] Error[9]: [graphShapeAnalyzer.cpp::throwIfError::1306] Error Code 9: Internal Error (Mul_650: broadcast dimensions must be conformable
)
[02/25/2022-00:00:24] [E] [TRT] ModelImporter.cpp:720: While parsing node number 650 [Mul -> "1310"]:
[02/25/2022-00:00:24] [E] [TRT] ModelImporter.cpp:721: --- Begin node ---
[02/25/2022-00:00:24] [E] [TRT] ModelImporter.cpp:722: input: "1307"
input: "2405"
output: "1310"
name: "Mul_650"
op_type: "Mul"

[02/25/2022-00:00:24] [E] [TRT] ModelImporter.cpp:723: --- End node ---
[02/25/2022-00:00:24] [E] [TRT] ModelImporter.cpp:726: ERROR: ModelImporter.cpp:179 In function parseGraph:
[6] Invalid Node - Mul_650
[graphShapeAnalyzer.cpp::throwIfError::1306] Error Code 9: Internal Error (Mul_650: broadcast dimensions must be conformable
)

The error indicates an invalid broadcast axis from the mul layer.
Would you mind double-checking if all the layers within your model are supported by the TensorRT?

Thanks

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.