search

NVIDIA / nv-wavenet

Reference implementation of real-time autoregressive wavenet inference
BSD 3-Clause "New" or "Revised" License
735 stars 126 forks source link

GPUassert: invalid device function ../nv_wavenet_util.cuh 48 #12

Closed PetrochukM closed 6 years ago

PetrochukM commented 6 years ago

Issue description

Error running nv-wavenet test for PyTorch

Code example

$ python3.6 nv-wavenet/pytorch/nv_wavenet_test.py
GPUassert: invalid device function ../nv_wavenet_util.cuh 48

System Info

PyTorch version: 0.4.0 Python version: 3.6.4 CUDA used to build PyTorch: 9.0.176

OS: Ubuntu 16.04.4 LTS GCC version: (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609 CMake version: version 3.5.1

Python version: 3.6 Is CUDA available: Yes CUDA runtime version: 9.0.176 GPU models and configuration: GPU 0: GeForce GTX 1080 Ti Nvidia driver version: 390.30 cuDNN version: Probably one of the following: /usr/lib/x86_64-linux-gnu/libcudnn.so.6.0.21 /usr/lib/x86_64-linux-gnu/libcudnn.so.7.0.5 /usr/local/lib/python2.7/dist-packages/torch/lib/libcudnn-7a90c013.so.7.0.5 /usr/local/lib/python3.5/dist-packages/torch/lib/libcudnn-3f9a723f.so.6.0.21

will-rice commented 6 years ago

I had a similar error the first time I tried the test. I updated to the newest version of CUDA and it fixed it. Also, check --arch parameter

PetrochukM commented 6 years ago

Ditto --arch parameter, discovered that before I saw your comment.