yolov10的onnx转tensorrt报错

[futureflsl]

调用：TensorRtSharp.Custom.Nvinfer.OnnxToEngine(@"C:\Users\Administrator\Desktop\yolov10n.onnx",1024);
报错：
[05/29/2024-07:28:53] [I] [TRT] [MemUsageChange] Init CUDA: CPU +523, GPU +0, now: CPU 8558, GPU 1190 (MiB)
[05/29/2024-07:28:54] [I] [TRT] [MemUsageChange] Init builder kernel library: CPU +240, GPU +68, now: CPU 8993, GPU 1258 (MiB)
[05/29/2024-07:28:54] [I] [TRT] ----------------------------------------------------------------
[05/29/2024-07:28:54] [I] [TRT] Input filename: C:\Users\Administrator\Desktop\yolov10n.onnx
[05/29/2024-07:28:54] [I] [TRT] ONNX IR version: 0.0.6
[05/29/2024-07:28:54] [I] [TRT] Opset version: 11
[05/29/2024-07:28:54] [I] [TRT] Producer name: pytorch
[05/29/2024-07:28:54] [I] [TRT] Producer version: 2.0.1
[05/29/2024-07:28:54] [I] [TRT] Domain:
[05/29/2024-07:28:54] [I] [TRT] Model version: 0
[05/29/2024-07:28:54] [I] [TRT] Doc string:
[05/29/2024-07:28:54] [I] [TRT] ----------------------------------------------------------------
[05/29/2024-07:28:54] [W] [TRT] onnx2trt_utils.cpp:369: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[05/29/2024-07:28:54] [I] [TRT] No importer registered for op: Mod. Attempting to import as plugin.
[05/29/2024-07:28:54] [I] [TRT] Searching for plugin: Mod, plugin_version: 1, plugin_namespace:
[05/29/2024-07:28:54] [E] [TRT] ModelImporter.cpp:773: While parsing node number 300 [Mod -> "/model.23/Mod_output_0"]:
[05/29/2024-07:28:54] [E] [TRT] ModelImporter.cpp:774: --- Begin node ---
[05/29/2024-07:28:54] [E] [TRT] ModelImporter.cpp:775: input: "/model.23/TopK_1_output_1"
input: "/model.23/Constant_12_output_0"
output: "/model.23/Mod_output_0"
name: "/model.23/Mod"
op_type: "Mod"
attribute {
name: "fmod"
i: 0
type: INT
}

[05/29/2024-07:28:54] [E] [TRT] ModelImporter.cpp:776: --- End node ---
[05/29/2024-07:28:54] [E] [TRT] ModelImporter.cpp:779: ERROR: builtin_op_importers.cpp:4890 In function importFallbackPluginImporter:
[8] Assertion failed: creator && "Plugin not found, are the plugin name, version, and namespace correct?"
load error: Assertion failed: creator && "Plugin not found, are the plugin name, version, and namespace correct?"
tensorRT load mask onnx model successfully!!!...
[05/29/2024-07:28:54] [E] [TRT] 4: [network.cpp::nvinfer1::Network::validate::2671] Error Code 4: Internal Error (Network must have at least one output)
try to save engine file now~~~
“WindowsFormsApp1.exe”(CLR v4.0.30319: WindowsFormsApp1.exe): 已加载“C:\Windows\Microsoft.Net\assembly\GAC_MSIL\mscorlib.resources\v4.0_4.0.0.0_zh-Hans_b77a5c561934e089\mscorlib.resources.dll”。模块已生成，不包含符号。
“System.AccessViolationException”类型的未经处理的异常在 TensorRtSharp.dll 中发生
尝试读取或写入受保护的内存。这通常指示其他内存已损坏。

尝试用官方yolov10命令直接导出tensorrt可以导出成功，但是用该项目无法调用

测试环境：
tensorrt8.6.1.6
cuda11.7+cudnn8.8.0

[guojin-yan]

你这个应该是模型导出onnx设置有问题，有问题可以入群交流，QQ群：945057948