Closed christopi closed 4 months ago
This error is because your tensorrt backend for onnxruntime is not installed, so onnxruntime fails and switches to its default backend (which is very, very slow), a straightforward solution is to comment out TensorrtExecutionProvider in https://github.com/AiuniAI/Unique3D/blob/ 4263b8c836950babedd9c8b6769aa7c41afa9dbc/scripts/load_onnx.py#L5 , and keep only CUDAExecutionProvider.
When I run the project on my ubuntu server, this error occurred and its generation time is quite slow over 10 mins.