Error converting onnx to TensorRT engine on Xavier

AI & Data Science Deep Learning (Training & Inference) TensorRT
,
1

I have [SSD MobileNet V2 FPNLite 640x640 model which I successfully converted to onnx using the following command :
python -m tf2onnx.convert --opset 11 --fold_const --inputs input_tensor:0[1,-1,-1,3] --saved-model C:\Users\RT\Desktop\inference_graph\saved_model --output C:\Users\RT\Desktop\inference_graph\saved_model\model1.onnx

Now on Nvidia Xavier with TensorRT 8.1 ,Im trying to convert the onnx model to tensorrt engine using the below command :
trtexec --onnx=/home/RT/Desktop/model111.onnx --saveEngine=/home/RT/Desktop/model111.trt

Im getting the below error :
[12/09/2021-16:53:41] [E] [TRT] ModelImporter.cpp:720: While parsing node number 3529 [TopK → “StatefulPartitionedCall/Postprocessor/BatchMultiClassNonMaxSuppression/MultiClassNonMaxSuppression/SortByField/TopKV2:0”]:
[12/09/2021-1I
6:53:41] [E] [TRT] ModelImporter.cpp:721: — Begin node —
[12/09/2021-16:53:41] [E] [TRT] ModelImporter.cpp:722: input: “StatefulPartitionedCall/Postprocessor/BatchMultiClassNonMaxSuppression/MultiClassNonMaxSuppression/Concatenate/concat_1:0”
input: “Unsqueeze__4691:0”
output: “StatefulPartitionedCall/Postprocessor/BatchMultiClassNonMaxSuppression/MultiClassNonMaxSuppression/SortByField/TopKV2:0”
output: “StatefulPartitionedCall/Postprocessor/BatchMultiClassNonMaxSuppression/MultiClassNonMaxSuppression/SortByField/TopKV2:1”
name: “StatefulPartitionedCall/Postprocessor/BatchMultiClassNonMaxSuppression/MultiClassNonMaxSuppression/SortByField/TopKV2”
op_type: “TopK”
attribute {
name: “sorted”
i: 1
type: INT
}

[12/09/2021-16:53:41] [E] [TRT] ModelImporter.cpp:723: — End node —
[12/09/2021-16:53:41] [E] [TRT] ModelImporter.cpp:726: ERROR: builtin_op_importers.cpp:4292 In function importTopK:
[8] Assertion failed: (inputs.at(1).is_weights()) && “This version of TensorRT only supports input K as an initializer.”
[12/09/2021-16:53:41] [E] Failed to parse onnx file
[12/09/2021-16:53:41] [I] Finish parsing network model
[12/09/2021-16:53:41] [E] Parsing model failed
[12/09/2021-16:53:41] [E] Engine creation failed

Appreciate any help
Thanks
model111.onnx (12.6 MB)

2

I moved this topic to the TensorRT sub-category to help get attention to this question.
Thanks

3

Hi,
Request you to share the ONNX model and the script if not shared already so that we can assist you better.
Alongside you can try few things:

  1. validating your model with the below snippet

check_model.py

import sys
import onnx
filename = yourONNXmodel
model = onnx.load(filename)
onnx.checker.check_model(model).
2) Try running your model with trtexec command.
https://github.com/NVIDIA/TensorRT/tree/master/samples/opensource/trtexec
In case you are still facing issue, request you to share the trtexec “”–verbose"" log for further debugging
Thanks!