Hi,
thank you for your great work.
I need some help when running the EdgeSam model.
if I run the model on CPU, it was OK.
But if I set param.providers[1].deviceType = 1;,
I can not get the right result.
I use onnxruntime 1.17, 1.9, or 1.23,CUDA is 12.9, and cuDNN is 9.5.
the error msg is like :
2025-11-26 17:26:36.0450088 [W:onnxruntime:, session_state.cc:1166 onnxruntime::VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf. 2025-11-26 17:26:36.0508138 [W:onnxruntime:, session_state.cc:1168 onnxruntime::VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments. Invalid handle. Cannot load symbol cudnnCreate
Any help is needed!
thanks again!
yours,
YL
Hi,
thank you for your great work.
I need some help when running the EdgeSam model.
if I run the model on CPU, it was OK.
But if I set
param.providers[1].deviceType = 1;,I can not get the right result.
I use onnxruntime 1.17, 1.9, or 1.23,CUDA is 12.9, and cuDNN is 9.5.
the error msg is like :
2025-11-26 17:26:36.0450088 [W:onnxruntime:, session_state.cc:1166 onnxruntime::VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf. 2025-11-26 17:26:36.0508138 [W:onnxruntime:, session_state.cc:1168 onnxruntime::VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments. Invalid handle. Cannot load symbol cudnnCreateAny help is needed!
thanks again!
yours,
YL