c++ gpu推理报错onnx runtime

gpu推理报错

缺少：cudnn_heuristic64_9.dll

Non-zero status code returned while running Conv node. Name:'Conv_0' Status Message: Failed to initialize CUDNN FrontendE:\_work\1\s\onnxruntime\core\providers\cuda\cudnn_fe_call.cc:99 onnxruntime::CudaCall E:\_work\1\s\onnxruntime\core\providers\cuda\cudnn_fe_call.cc:91 onnxruntime::CudaCall CUDNN_FE failure 11: CUDNN_BACKEND_API_FAILED ; GPU=0 ; hostname=DESK-SAWAOI1 ; file=E:\_work\1\s\onnxruntime\core\providers\cuda\nn\conv.cc ; line=237 ; expr=s_.cudnn_fe_graph->build_operation_graph(handle);

with the cudnn frontend json:
{"context":{"compute_data_type":"FLOAT","intermediate_data_type":"FLOAT","io_data_type":"FLOAT","name":"","sm_count":-1},"cudnn_backend_version":"9.1.1","cudnn_frontend_version":11200,"json_version":"1.0","nodes":[{"compute_data_type":"FLOAT","dilation":[1,1],"inputs":{"W":"w","X":"x"},"math_mode":"CROSS_CORRELATION","name":"0","outputs":{"Y":"0::Y"},"post_padding":[2,2],"pre_padding":[2,2],"stride":[2,2],"tag":"CONV_FPROP"}],"tensors":{"0::Y":{"data_type":"FLOAT","dim":[1,32,320,320],"is_pass_by_value":false,"is_virtual":false,"name":"0::Y","pass_by_value":null,"reordering_type":"NONE","stride":[3276800,102400,320,1],"uid":0,"uid_assigned":false},"w":{"data_type":"FLOAT","dim":[32,3,6,6],"is_pass_by_value":false,"is_virtual":false,"name":"w","pass_by_value":null,"reordering_type":"NONE","stride":[108,36,6,1],"uid":0,"uid_assigned":false},"x":{"data_type":"FLOAT","dim":[1,3,640,640],"is_pass_by_value":false,"is_virtual":false,"name":"x","pass_by_value":null,"reordering_type":"NONE","stride":[1228800,409600,640,1],"uid":1,"uid_assigned":true}}}

 

缺少：

cublas64_12.dll

OrtSessionOptionsAppendExecutionProvider_Cuda: Failed to load shared library

 

gt1030 gpu不支持，建议rtx3060

Non-zero status code returned while running QuickGelu node. Name:'Mul_2/QuickGeluFusion/' Status Message: CUDA error cudaErrorNoKernelImageForDevice:no kernel image is available for execution on the device