火炬未在启用CUDA的情况下编译

Question

我如何安装PyTorch：

conda install pytorch torchvision cudatoolkit=10.1 -c pytorch

我如何安装Tensor Flow：

pip install tensorflow

CUDA工具包已安装并通过以下方式进行检查：

nvcc --version

按照说明将cuDNN 7.6.4 SDK导入CUDA文件夹，示例如下：

NVIDIDA卡和驱动程序信息

=====================================

代码：

In [1]:
from __future__ import print_function
from __future__ import division
import torch
import torch.nn as nn
import torch.optim as optim
from torch.utils.data import DataLoader, TensorDataset, Dataset
from torch import Tensor
import numpy as np
import pandas
import os
from PIL import Image
import glob
from os import listdir
from os.path import isfile, join
import torchvision
from torchvision import datasets, models, transforms
import matplotlib.pyplot as plt
import time
import math
import copy
import torch.utils.data
import cv2
import tensorflow as tf
print("PyTorch Version: ",torch.__version__)
print("Torchvision Version: ",torchvision.__version__)
print("Tensor Flow Version: ", tf.__version__)
tf.test.gpu_device_name()

Out [1]:
PyTorch Version:  1.6.0
Torchvision Version:  0.7.0
Tensor Flow Version:  2.3.0

'/device:GPU:0'

In [2]:
torch.cuda.current_device()

Out [2]:
---------------------------------------------------------------------------
AssertionError                            Traceback (most recent call last)
<ipython-input-2-3380d2c12118> in <module>
----> 1 torch.cuda.current_device()

~\Anaconda3\lib\site-packages\torch\cuda\__init__.py in current_device()
382 def current_device() -> int:
383     r"""Returns the index of a currently selected device."""
--> 384     _lazy_init()
385     return torch._C._cuda_getDevice()
386 

~\Anaconda3\lib\site-packages\torch\cuda\__init__.py in _lazy_init()
184             raise RuntimeError(
185                 "Cannot re-initialize CUDA in forked subprocess. " + msg)
--> 186         _check_driver()
187         if _cudart is None:
188             raise AssertionError(

~\Anaconda3\lib\site-packages\torch\cuda\__init__.py in _check_driver()
59 def _check_driver():
60     if not hasattr(torch._C, '_cuda_isDriverSufficient'):
---> 61         raise AssertionError("Torch not compiled with CUDA enabled")
62     if not torch._C._cuda_isDriverSufficient():
63         if torch._C._cuda_getDriverVersion() == 0:

AssertionError: Torch not compiled with CUDA enabled

有人可以帮助我了解这里发生的事情吗？ 我已经卸载，然后重新安装了PyTorch和Tensor Flow几次，都没有成功。