mxnet-cu102无法调用GPU

我在成功安装CUDA10.2和其对应的CUDNN的情况下,用pip安装了mxnet-cu102,在使用时碰到了如下问题:

代码print(mx.gpu())能成功打印出gpu(0)

但是,
a = nd.array([1, 2, 3], ctx=mx.gpu(0))
print(a)
这段代码会之间陷入无响应状态,既不会输出结果也不会报错,但是GPU监视窗口显示有一个python进程。

以下是我的配置信息:
----------Python Info----------
Version : 3.6.13
Compiler : MSC v.1916 64 bit (AMD64)
Build : (‘default’, ‘Mar 16 2021 11:37:27’)
Arch : (‘64bit’, ‘WindowsPE’)
------------Pip Info-----------
Version : 21.2.2
Directory : G:\anaconda3\envs\gluon\lib\site-packages\pip
----------MXNet Info-----------
Version : 1.7.0
Directory : G:\anaconda3\envs\gluon\lib\site-packages\mxnet
Commit hash file “G:\anaconda3\envs\gluon\lib\site-packages\mxnet\COMMIT_HASH” not found. Not installed from pre-built package or built from source.
Library : [‘G:\anaconda3\envs\gluon\lib\site-packages\mxnet\libmxnet.dll’]
Build features:
:heavy_check_mark: CUDA
:heavy_check_mark: CUDNN
:heavy_multiplication_x: NCCL
:heavy_check_mark: CUDA_RTC
:heavy_multiplication_x: TENSORRT
:heavy_multiplication_x: CPU_SSE
:heavy_multiplication_x: CPU_SSE2
:heavy_multiplication_x: CPU_SSE3
:heavy_multiplication_x: CPU_SSE4_1
:heavy_multiplication_x: CPU_SSE4_2
:heavy_multiplication_x: CPU_SSE4A
:heavy_multiplication_x: CPU_AVX
:heavy_multiplication_x: CPU_AVX2
:heavy_check_mark: OPENMP
:heavy_multiplication_x: SSE
:heavy_multiplication_x: F16C
:heavy_multiplication_x: JEMALLOC
:heavy_check_mark: BLAS_OPEN
:heavy_multiplication_x: BLAS_ATLAS
:heavy_multiplication_x: BLAS_MKL
:heavy_multiplication_x: BLAS_APPLE
:heavy_check_mark: LAPACK
:heavy_multiplication_x: MKLDNN
:heavy_check_mark: OPENCV
:heavy_multiplication_x: CAFFE
:heavy_multiplication_x: PROFILER
:heavy_multiplication_x: DIST_KVSTORE
:heavy_multiplication_x: CXX14
:heavy_multiplication_x: INT64_TENSOR_SIZE
:heavy_check_mark: SIGNAL_HANDLER
:heavy_multiplication_x: DEBUG
:heavy_multiplication_x: TVM_OP
----------System Info----------
Platform : Windows-10-10.0.22000-SP0
system : Windows
node : txh
release : 10
version : 10.0.22000
----------Hardware Info----------
machine : AMD64
processor : Intel64 Family 6 Model 141 Stepping 1, GenuineIntel
Name
11th Gen Intel® Core™ i7-11800H @ 2.30GHz