Invalid node error when generating model engine using the script install_ess_models.sh

[arvindrk2]

I am trying to generate the model engine from the onnx file on my jetson orin nano using the install_ess_models.sh script. I am using the model version v4.1.0_onnx. But I keep running into the error shown below. The issue occurs when I try to run trtexec separately in the terminal as well. Can someone help me understand why I am running into this and how I can fix it?

[06/13/2025-19:41:36] [E] [TRT] parsers/onnx/ModelImporter.cpp:961: While parsing node number 267 [Conv -> "/model/prediction/layers/layers.0/Conv_output_0"]:
[06/13/2025-19:41:36] [E] [TRT] parsers/onnx/ModelImporter.cpp:964: --- Begin node ---
input: "/model/prediction/Reshape_output_0"
input: "model.prediction.layers.0.weight"
input: "model.prediction.layers.0.bias"
output: "/model/prediction/layers/layers.0/Conv_output_0"
name: "/model/prediction/layers/layers.0/Conv"
op_type: "Conv"
attribute {
name: "dilations"
ints: 1
type: INTS
}
attribute {
name: "group"
i: 1
type: INT
}
attribute {
name: "kernel_shape"
ints: 1
type: INTS
}
attribute {
name: "pads"
ints: 0
ints: 0
type: INTS
}
attribute {
name: "strides"
ints: 1
type: INTS
}
[06/13/2025-19:41:36] [E] [TRT] parsers/onnx/ModelImporter.cpp:965: --- End node ---
[06/13/2025-19:41:36] [E] [TRT] parsers/onnx/ModelImporter.cpp:967: ERROR: parsers/onnx/ModelImporter.cpp:202 In function parseNode:
[6] Invalid Node - /model/prediction/layers/layers.0/Conv
[06/13/2025-19:41:36] [E] Failed to parse onnx file
[06/13/2025-19:41:36] [I] Finished parsing network model. Parse time: 0.39815
[06/13/2025-19:41:36] [E] Parsing model failed
[06/13/2025-19:41:36] [E] Failed to create engine from model or file.
[06/13/2025-19:41:36] [E] Engine set up failed

[jaiveersinghNV]

What JetPack version and what TensorRT version do you have installed?

Additionally, can you inspect the ONNX file in a tool like this one and make sure the file was valid:
https://netron.app/

[arvindrk2]

What JetPack version and what TensorRT version do you have installed?

Additionally, can you inspect the ONNX file in a tool like this one and make sure the file was valid: https://netron.app/

@jaiveersinghNV Thank you for the response. Here are the spec you asked for,
Jetpack 6.2
TensorRT 10.3.0

I inspected it with netron and I tried using graphsurgeon to fix it but it led to more errors. I also checked the onnx file I downloaded from the NGC catalog using onnxruntime checker and I got the following error

/.local/lib/python3.10/site-packages/onnx/checker.py", line 164, in check_model
C.check_model_path(
onnx.onnx_cpp2py_export.checker.ValidationError: No Op registered for FusedConcatConv3x3 with domain_version of 13
==> Context: Bad node spec for node. Name: FusedConcatConv3x3_left OpType: FusedConcatConv3x3

But the onnx file wasn't exported by me, I just downloaded it directly from the NGC catalog. So that has me thinking that there could be something fundamentally wrong in the way I using the files or my system specs don't support it. Although, the catalog file description does mention that it should work on my system (Quoted below).
"ESS is designed and tested to run on NVIDIA hardware and software. For hardware, the model can run on any NVIDIA GPU including NVIDIA Jetson devices. For software, the model is intended for Ubuntu 22.04 with CUDA 12.2 or Jetpack 6.0 and later. This model can only be used with TensorRT."
Please let me know if you need any more details. I would like to have this issue resolved so any help would be greatly appreciated.

[jaiveersinghNV]

The installed software looks current, so I suspect there's an issue with the copy of the file you have downloaded.

I just downloaded a fresh copy of the asset from the NGC Model Card. Here are the md5sums of the relevant files:

- 76971095f27bc006440d928ddd174f22  dnn_stereo_disparity_v4.1.0_onnx.tar.gz

(after uncompressing)
- 70ea55dc243ae9a7fc2757123867a23c  dnn_stereo_disparity_v4.1.0_onnx/ess.onnx
- 770996272facf8950841028587be3116  dnn_stereo_disparity_v4.1.0_onnx/light_ess.onnx

Can you clarify if the graph correctly rendered in Netron? I am able to view both these files with no errors or warnings in Netron.

[arvindrk2]

The installed software looks current, so I suspect there's an issue with the copy of the file you have downloaded.

I just downloaded a fresh copy of the asset from the NGC Model Card. Here are the md5sums of the relevant files:

- 76971095f27bc006440d928ddd174f22  dnn_stereo_disparity_v4.1.0_onnx.tar.gz

(after uncompressing)
- 70ea55dc243ae9a7fc2757123867a23c  dnn_stereo_disparity_v4.1.0_onnx/ess.onnx
- 770996272facf8950841028587be3116  dnn_stereo_disparity_v4.1.0_onnx/light_ess.onnx

Can you clarify if the graph correctly rendered in Netron? I am able to view both these files with no errors or warnings in Netron.

Oh, yes. The graph shows up perfectly in my browser with no errors or warnings. I can even see the node (and its attributes) that is supposedly causing the error during the trtexec conversion. So initially, I had a few nodes that had an error. Then, I tried the trtexec conversion while loading the aarch64 plugin that is included in the files. This fixes all other errors except this one.