30系列显卡可以用了吗

中文版 mxnet
1

看到有mxnet-cu112的下载，是支持30系列显卡的吗?
具体怎么安装

2

到这里我就不会搞了
OSError: libcudnn.so.8: cannot open shared object file: No such file or directory

import mxnet as mx
File “/home/sgdxhhw/.conda/envs/mxnet/lib/python3.6/site-packages/mxnet/base.py”, line 356, in
_LIB = _load_lib()
File “/home/sgdxhhw/.conda/envs/mxnet/lib/python3.6/site-packages/mxnet/base.py”, line 347, in _load_lib
lib = ctypes.CDLL(lib_path[0], ctypes.RTLD_LOCAL)
File “/home/sgdxhhw/.conda/envs/mxnet/lib/python3.6/ctypes/init.py”, line 348, in init
self._handle = _dlopen(self._name, mode)
OSError: libcudnn.so.8: cannot open shared object file: No such file or directory

3

我用的包

Anaconda3-2021.11-Linux-x86_64.sh
mxnet_cu112-1.9.0-py3-none-manylinux2014_x86_64.whl
NVIDIA-Linux-x86_64-510.60.02.run
cuda_11.2.0_460.27.04_linux.run

变量 ~/.bashrc

ulimit -c unlimited
ulimit -s 819200

export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-11.2/lib64
export PATH=$PATH:/usr/local/cuda-11.2/bin
export CUDA_HOME=$CUDA_HOME:/usr/local/cuda-11.2
export PATH="/home/sgdxhhw/Anaconda3/bin:$PATH"

4

可以用了 不过还是有错误
我是安装下面这些
python==3.8
conda install nccl -c conda-forge

pip install autogluon

conda install pytorch=1.7.1 torchvision cudatoolkit=11.0 cudnn -c pytorch -c conda-forge -c nvidia

错误：
Python 3.8.13
(mx) sgdxhhw@sgdxhhw-desktop:~$ python
Python 3.8.13 (default, Mar 28 2022, 11:38:47)
[GCC 7.5.0] :: Anaconda, Inc. on linux
Type “help”, “copyright”, “credits” or “license” for more information.

import mxnet as mx
a= mx.nd.array([1,2,3])
a

[1. 2. 3.]
<NDArray 3 @cpu(0)>

b= mx.nd.array([1,2,3],mx.gpu())
[19:45:29] …/src/base.cc:79: cuDNN lib mismatch: linked-against version 8201 != compiled-against version 8100. Set MXNET_CUDNN_LIB_CHECKING=0 to quiet this warning.
b

[1. 2. 3.]
<NDArray 3 @gpu(0)>