Convert onnx to engine

[sctrueew]

Hi, thanks for your work. when I want to convert onnx to engine I gt this error:
I convert yolov5s-seg to onnx first then I use your converter

/18/2022-19:10:17] [W] [TRT] onnx2trt_utils.cpp:366: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[11/18/2022-19:10:17] [E] [TRT] ModelImporter.cpp:769: While parsing node number 142 [Resize -> "onnx::Concat_277"]:
[11/18/2022-19:10:17] [E] [TRT] ModelImporter.cpp:770: --- Begin node ---
[11/18/2022-19:10:17] [E] [TRT] ModelImporter.cpp:771: input: "onnx::Resize_272"
input: "onnx::Resize_276"
input: "onnx::Resize_474"
output: "onnx::Concat_277"
name: "Resize_142"
op_type: "Resize"
attribute {
  name: "coordinate_transformation_mode"
  s: "asymmetric"
  type: STRING
}
attribute {
  name: "cubic_coeff_a"
  f: -0.75
  type: FLOAT
}
attribute {
  name: "mode"
  s: "nearest"
  type: STRING
}
attribute {
  name: "nearest_mode"
  s: "floor"
  type: STRING
}

[11/18/2022-19:10:17] [E] [TRT] ModelImporter.cpp:772: --- End node ---
[11/18/2022-19:10:17] [E] [TRT] ModelImporter.cpp:775: ERROR: builtin_op_importers.cpp:3603 In function importResize:
[8] Assertion failed: scales.is_weights() && "Resize scales must be an initializer!"
Assertion failed: scales.is_weights() && "Resize scales must be an initializer!"
successfully load the onnx model
[11/18/2022-19:11:33] [E] [TRT] 4: [network.cpp::nvinfer1::Network::validate::2605] Error Code 4: Internal Error (Network must have at least one output)

[fish-kong]

Is your tensorrt version 7. x or lower? please try to use 8.x version. this error occured when i used 7.x last few years，because 7.x do not support upsample(torch)