check_cudnn_acceptable_type？烦请大佬解惑!!
0
[待解决问题]
最近我老板让我跑一个深度学习的代码，老板可能忘了我还是什么都不懂的小白···
在配置了n天的环境后，cuda的环境貌似是好了···
但跑的时候仍然报错，而这报的是CUDNN，可是我cudnn是安装好了的···
网上查找也找不到任何相关的···
实在不知道怎么办了···发出了第一帖，求求大佬们帮忙看看！！谢谢！！
贴出报错
1 D:\Anaconda2\lib\site-packages\h5py\__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`. 2 from ._conv import register_converters as _register_converters 3 Traceback (most recent call last): 4 File "D:\Documents\ddnn\examples\test.py", line 8, in <module> 5 from elaas.elaas import Collection 6 File "..\elaas\elaas.py", line 6, in <module> 7 from .family.simple import SimpleHybridFamily 8 File "..\elaas\family\simple.py", line 1, in <module> 9 from chainer_sequential.chain import Chain 10 File "..\chainer_sequential\__init__.py", line 1, in <module> 11 import link 12 File "..\chainer_sequential\link.py", line 3, in <module> 13 import weightnorm 14 File "..\chainer_sequential\weightnorm\__init__.py", line 3, in <module> 15 from deconvolution_2d import Deconvolution2D 16 File "..\chainer_sequential\weightnorm\deconvolution_2d.py", line 20, in <module> 17 _check_cudnn_acceptable_type = convolution_2d._check_cudnn_acceptable_type 18 AttributeError: 'module' object has no attribute '_check_cudnn_acceptable_type' 19 [Finished in 13.4s with exit code 1]
deconvolution_2d. 1 import math
2 import numpy as np 3 from six import moves 4 from chainer import cuda, Variable, initializers, link, function 5 from chainer.utils import conv, type_check 6 from chainer.functions.connection import deconvolution_2d, convolution_2d 7 8 if cuda.cudnn_enabled: 9 cudnn = cuda.cudnn 10 libcudnn = cuda.cudnn.cudnn 11 _cudnn_version = libcudnn.getVersion() 12 _fwd_pref = libcudnn.CUDNN_CONVOLUTION_FWD_SPECIFY_WORKSPACE_LIMIT 13 if _cudnn_version >= 4000: 14 _bwd_filter_pref = \ 15 libcudnn.CUDNN_CONVOLUTION_BWD_FILTER_SPECIFY_WORKSPACE_LIMIT 16 _bwd_data_pref = \ 17 libcudnn.CUDNN_CONVOLUTION_BWD_DATA_SPECIFY_WORKSPACE_LIMIT 18 19 _check_cudnn_acceptable_type = convolution_2d._check_cudnn_acceptable_type
##报错位置
