Hello.
I installed torch as per getting started locally guide. I have a dual xeon system with RTX 3080Ti and CUDA 11.6, CUDNN8.4.1 … They are all working fine. However with torch I get error
NVIDIA GeForce RTX 3080 Ti with CUDA capability sm_86 is not compatible with the current PyTorch installation.
The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_70.
If you want to use the NVIDIA GeForce RTX 3080 Ti GPU with PyTorch, please check the instructions at Start Locally | PyTorch
I have already followed the instructions but there is no success.
The whole steps are as under:
metanet@metanet-Precision-Tower-7810:~/ProgramFiles/speechbrain/recipes$ pip3 install torch torchvision torchaudio --extra-index-url
Defaulting to user installation because normal site-packages is not writeable
Looking in indexes: https//pypiorg/simple, https//pypingcnvidiacom, https//downloadpytorchorg/whl/cu116
Requirement already satisfied: torch in /home/metanet/.local/lib/python3.8/site-packages (1.11.0)
Requirement already satisfied: torchvision in /home/metanet/.local/lib/python3.8/site-packages (0.12.0)
Requirement already satisfied: torchaudio in /home/metanet/.local/lib/python3.8/site-packages (0.11.0)
Requirement already satisfied: typing-extensions in /home/metanet/.local/lib/python3.8/site-packages (from torch) (4.2.0)
Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /home/metanet/.local/lib/python3.8/site-packages (from torchvision) (9.1.1)
Requirement already satisfied: numpy in /home/metanet/.local/lib/python3.8/site-packages (from torchvision) (1.21.6)
Requirement already satisfied: requests in /usr/lib/python3/dist-packages (from torchvision) (2.22.0)
metanet@metanet-Precision-Tower-7810:~/ProgramFiles/speechbrain/recipes$ nvidia-smi
Sat Jul 2 11:24:12 2022
±----------------------------------------------------------------------------+
| NVIDIA-SMI 510.47.03 Driver Version: 510.47.03 CUDA Version: 11.6 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|===============================+======================+======================|
| 0 NVIDIA GeForce … On | 00000000:03:00.0 On | N/A |
| 30% 45C P8 32W / 350W | 494MiB / 12288MiB | 13% Default |
| | | N/A |
±------------------------------±---------------------±---------------------+
±----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| 0 N/A N/A 2298 G /usr/lib/xorg/Xorg 190MiB |
| 0 N/A N/A 2713 G /usr/bin/gnome-shell 51MiB |
| 0 N/A N/A 3746 G …257163255691216877,131072 81MiB |
| 0 N/A N/A 23276 G …veSuggestionsOnlyOnDemand 92MiB |
| 0 N/A N/A 24491 G …b/thunderbird/thunderbird 74MiB |
±----------------------------------------------------------------------------+
**metanet@metanet-Precision-Tower-7810:~/ProgramFiles/speechbrain/recipes$ cat check-device.py **
import torch
torch.cuda.is_available()
torch.cuda.current_device()
torch.cuda.device(0)
torch.cuda.device_count()
torch.cuda.get_device_name(0)
device = torch.device(‘cuda’ if torch.cuda.is_available() else ‘cpu’)
print(‘Using device:’, device)
print()
if device.type == ‘cuda’:
print(torch.cuda.get_device_name(0))
print(‘Memory Usage:’)
print(‘Allocated:’, round(torch.cuda.memory_allocated(0)/10243,1), ‘GB’)
print('Cached: ', round(torch.cuda.memory_reserved(0)/10243,1), ‘GB’)
$ python3 check-device.py
/home/metanet/.local/lib/python3.8/site-packages/torch/cuda/init.py:145: UserWarning:
NVIDIA GeForce RTX 3080 Ti with CUDA capability sm_86 is not compatible with the current PyTorch installation.
The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_70.
If you want to use the NVIDIA GeForce RTX 3080 Ti GPU with PyTorch, please check the instructions at Start Locally | PyTorch
warnings.warn(incompatible_device_warn.format(device_name, capability, " ".join(arch_list), device_name))
Using device: cuda
NVIDIA GeForce RTX 3080 Ti
Memory Usage:
Allocated: 0.0 GB
Cached: 0.0 GB