我正在尝试这些给定的说明https://developer.ibm.com/tutorials/install-tensorflow-on-power/ 但出现以下错误。我知道也有类似的问题，但它们没有类似的错误。
问题说明 在电源体系结构计算机上从源构建张量流失败。
错误再现性 按照说明given here在电源体系结构系统上从源代码构建和安装tensorflow。 IBM POWER8 CPU，NVIDIA Tesla P100 GPU
错误消息
WARNING: Output base '/autofs/nccs-svm1_home1/shubhankar/.cache/bazel/_bazel_shubhankar/e74c0b81a4afde0b86e7c039eda6fc5e' is on NFS. This may lead to surprising failures and undetermined behavior.
ERROR: Skipping '//tensorflow/tools/pip_package:build_pip_package': error loading package 'tensorflow/tools/pip_package': Encountered error while reading extension file 'cuda/build_defs.bzl': no such package '@local_config_cuda//cuda': Traceback (most recent call last):
File "/autofs/nccs-svm1_home1/shubhankar/tensorflow/third_party/gpus/cuda_configure.bzl", line 1166
_create_local_cuda_repository(repository_ctx)
File "/autofs/nccs-svm1_home1/shubhankar/tensorflow/third_party/gpus/cuda_configure.bzl", line 995, in _create_local_cuda_repository
_get_cuda_config(repository_ctx)
File "/autofs/nccs-svm1_home1/shubhankar/tensorflow/third_party/gpus/cuda_configure.bzl", line 750, in _get_cuda_config
_cudnn_version(repository_ctx, cudnn_install_base..., ...)
File "/autofs/nccs-svm1_home1/shubhankar/tensorflow/third_party/gpus/cuda_configure.bzl", line 464, in _cudnn_version
_find_cudnn_header_dir(repository_ctx, cudnn_install_base...)
File "/autofs/nccs-svm1_home1/shubhankar/tensorflow/third_party/gpus/cuda_configure.bzl", line 707, in _find_cudnn_header_dir
auto_configure_fail(("Cannot find cudnn.h under %s" ...))
File "/autofs/nccs-svm1_home1/shubhankar/tensorflow/third_party/gpus/cuda_configure.bzl", line 210, in auto_configure_fail
fail(("\n%sCuda Configuration Error:%...)))
Cuda Configuration Error: Cannot find cudnn.h under /autofs/nccs-svm1_home1/shubhankar/libcudnn7_7.0.3.11-1+cuda8.0_ppc64el/lib/powerpc64le-linux-gnu
WARNING: Target pattern parsing failed.
ERROR: error loading package 'tensorflow/tools/pip_package': Encountered error while reading extension file 'cuda/build_defs.bzl': no such package '@local_config_cuda//cuda': Traceback (most recent call last):
File "/autofs/nccs-svm1_home1/shubhankar/tensorflow/third_party/gpus/cuda_configure.bzl", line 1166
_create_local_cuda_repository(repository_ctx)
File "/autofs/nccs-svm1_home1/shubhankar/tensorflow/third_party/gpus/cuda_configure.bzl", line 995, in _create_local_cuda_repository
_get_cuda_config(repository_ctx)
File "/autofs/nccs-svm1_home1/shubhankar/tensorflow/third_party/gpus/cuda_configure.bzl", line 750, in _get_cuda_config
_cudnn_version(repository_ctx, cudnn_install_base..., ...)
File "/autofs/nccs-svm1_home1/shubhankar/tensorflow/third_party/gpus/cuda_configure.bzl", line 464, in _cudnn_version
_find_cudnn_header_dir(repository_ctx, cudnn_install_base...)
File "/autofs/nccs-svm1_home1/shubhankar/tensorflow/third_party/gpus/cuda_configure.bzl", line 707, in _find_cudnn_header_dir
auto_configure_fail(("Cannot find cudnn.h under %s" ...))
File "/autofs/nccs-svm1_home1/shubhankar/tensorflow/third_party/gpus/cuda_configure.bzl", line 210, in auto_configure_fail
fail(("\n%sCuda Configuration Error:%...)))
Cuda Configuration Error: Cannot find cudnn.h under /autofs/nccs-svm1_home1/shubhankar/libcudnn7_7.0.3.11-1+cuda8.0_ppc64el/lib/powerpc64le-linux-gnu
INFO: Elapsed time: 0.197s
FAILED: Build did NOT complete successfully (0 packages loaded)
currently loading: tensorflow/tools/pip_package
系统配置 操作系统：Linux版本3.10.0-693.21.1.el7.ppc64le（mockbuild@ppc-053.build.eng.bos.redhat.com）（gcc版本4.8.5 20150623（Red Hat 4.8.5-16）（GCC ）） Python版本：3.5.6 CUDA / cuDNN版本：8.0 / 7.1.4
GPU型号和配置：NVIDIA P100 GCC版本（如果从源代码编译）：4.8.5 CMake版本：3.12.0 其他任何相关库的版本：Spectrum_mpi而不是libopenmpi-dev-cuda
openjdk版本“ 1.8.0_161” OpenJDK运行时环境（内部版本1.8.0_161-b14） OpenJDK 64位服务器VM（内部版本25.161-b14，混合模式）
没有sudo访问权限，因此也不能使用apt-get，也不能修改/ usr / local / *但必须安装所有依赖项。