-
Notifications
You must be signed in to change notification settings - Fork 284
Description
how to fix this ?
2025-04-08 15:59:31.4518965 [E:onnxruntime:Default, tensorrt_execution_provider.h:89 onnxruntime::TensorrtLogger::log] [2025-04-08 07:59:31 ERROR] [virtualMemoryBuffer.cpp::nvinfer1::StdVirtualMemoryBufferImpl::resizePhysical::141] Error Code 1: Cuda Driver (invalid argument)
2025-04-08 15:59:31.4999700 [E:onnxruntime:Default, tensorrt_execution_provider.h:89 onnxruntime::TensorrtLogger::log] [2025-04-08 07:59:31 ERROR] [globWriter.cpp::nvinfer1::builder::HybridGlobWriter::HybridGlobWriter::497] Error Code 2: OutOfMemory (Requested size was 1024 bytes.)
2025-04-08 15:59:31.5034045 [E:onnxruntime:, sequential_executor.cc:572 onnxruntime::ExecuteKernel] Non-zero status code returned while running TRTKernel_graph_main_graph_927312078466753112_0 node. Name:'TensorrtExecutionProvider_TRTKernel_graph_main_graph_927312078466753112_0_0' Status Message: TensorRT EP failed to create engine from network.
EP Error: [ONNXRuntimeError] : 11 : EP_FAIL : Non-zero status code returned while running TRTKernel_graph_main_graph_927312078466753112_0 node. Name:'TensorrtExecutionProvider_TRTKernel_graph_main_graph_927312078466753112_0_0' Status Message: TensorRT EP failed to create engine from network. using ['TensorrtExecutionProvider', 'CUDAExecutionProvider', 'CPUExecutionProvider']
Falling back to ['CUDAExecutionProvider', 'CPUExecutionProvider'] and retrying.