../rtSafe/safeContext.cpp (133) - Cudnn Error in configure: 7 (CUDNN_STATUS_MAPPING_ERROR)

AI & Data Science Deep Learning (Training & Inference) TensorRT
1

[E] [TRT] …/rtSafe/safeContext.cpp (133) - Cudnn Error in configure: 7 (CUDNN_STATUS_MAPPING_ERROR)
[E] [TRT] FAILED_EXECUTION: std::exception

enviroment:xavier cuda-10.2 cudnn-7.6.3 tensorrt-6.0

2

Hi,

Could you please elaborate more on the use case and how TRT is used here?
Also, if possible could you please share the script and model files to reproduce the issue so we can better help?

Thanks

3

Hi,
I have same problem when I import pycuda.autoinit and foward model by torch2trt.
Example of my code like this:

*import pycuda.autoinit *
from torch2trt import TRTModule
model_trt = TRTModule()
model_trt.load_state_dict(torch.load(‘model_path’))
model_trt.forward(images)

Help me, please!

5

Excuse me, have you solved the problem yet ？