RuntimeError: CUDA environment is not correctly set up

[jessican521]

When I training the VoxResNet in slice_version，The error shows:
$ python train.py -g 0 -f dataset_train.json
Namespace(display_step=1000, gpu=0, input_file='dataset_train.json', iteration=10000, learning_rate=0.001, n_batch=1, out='vrn.npz', shape=[80, 80, 80], weight_decay=0.0005)
Traceback (most recent call last):
File "train.py", line 84, in
main()
File "train.py", line 51, in main
chainer.cuda.get_device(args.gpu).use()
File "/usr/local/lib/python2.7/dist-packages/chainer/cuda.py", line 222, in get_device
check_cuda_available()
File "/usr/local/lib/python2.7/dist-packages/chainer/cuda.py", line 87, in check_cuda_available
raise RuntimeError(msg)
RuntimeError: CUDA environment is not correctly set up
(see https://github.com/pfnet/chainer#installation).'module' object has no attribute 'get_build_version'

My cuda version is cuda8.0， cudnn is cudnn-8.0-linux-x64-v6.0 and the chainer is 1.23.0