HDF5库版本与windows x64最新Tensorflow内部版本不匹配错误

Question

由于Tensorflow 1.0.1版本中仍然存在bug，我安装了最新的夜间构建Tensorflow cpu

不幸的是，对于这个版本，现在内核崩溃的原因不同，我得到了这个错误：

Warning! ***HDF5 library version mismatched error***
The HDF5 header files used to compile this application do not match
the version used by the HDF5 library to which this application is linked.
Data corruption or segmentation faults may occur if the application continues.
This can happen when an application was compiled by one version of HDF5 but
linked with a different version of static or shared HDF5 library.
You should recompile the application or check your shared library related
settings such as 'LD_LIBRARY_PATH'.
You can, at your own risk, disable this warning by setting the environment
variable 'HDF5_DISABLE_VERSION_CHECK' to a value of '1'.
Headers are 1.8.15, library is 1.8.18
      SUMMARY OF THE HDF5 CONFIGURATION
      =================================

General Information:
-------------------
                   HDF5 Version: 1.8.18
                  Configured on: 2017-03-06
                  Configured by: Visual Studio 14 2015 Win64
                 Configure mode: CMAKE 3.7.2
                    Host system: Windows-10.0.14393
              Uname information: Windows
                       Byte sex: little-endian
                      Libraries:
             Installation point: C:/Program Files/HDF5

Compiling Options:
------------------
               Compilation Mode:
                     C Compiler: C:/Program Files (x86)/Microsoft Visual Studio 14.0/VC/bin/x86_amd64/cl.exe
                         CFLAGS: /DWIN32 /D_WINDOWS /W3
                      H5_CFLAGS:
                      AM_CFLAGS:
                       CPPFLAGS:
                    H5_CPPFLAGS:
                    AM_CPPFLAGS:
               Shared C Library: YES
               Static C Library: YES
  Statically Linked Executables: OFF
                        LDFLAGS: /machine:x64
                     AM_LDFLAGS:
                Extra libraries: X:inclib-vc14-x64/zlib.lib;X:/inclib-vc14-x64/libsz.lib;X:/inclib-vc14-x64/libaec.lib
                       Archiver:
                         Ranlib:
              Debugged Packages:
                    API Tracing: OFF

Languages:
----------
                        Fortran: OFF
               Fortran Compiler:
          Fortran 2003 Compiler:
                  Fortran Flags:
               H5 Fortran Flags:
               AM Fortran Flags:
         Shared Fortran Library: YES
         Static Fortran Library: YES

                            C++: ON
                   C++ Compiler: C:/Program Files (x86)/Microsoft Visual Studio 14.0/VC/bin/x86_amd64/cl.exe
                      C++ Flags: /DWIN32 /D_WINDOWS /W3 /GR /EHsc
                   H5 C++ Flags:
                   AM C++ Flags:
             Shared C++ Library: YES
             Static C++ Library: YES

Features:
---------
                  Parallel HDF5: OFF
             High Level library: ON
                   Threadsafety: OFF
            Default API Mapping: v18
 With Deprecated Public Symbols: ON
         I/O filters (external):  DEFLATE DECODE ENCODE
                            MPE:
                     Direct VFD:
                        dmalloc:
Clear file buffers before write: ON
           Using memory checker: OFF
         Function Stack Tracing: OFF
      Strict File Format Checks: OFF

脚本崩溃了whentf.contrib.layers时.卷积2d被调用

我使用的是Windows10x64，并且我已经安装了为其他库发布的所有可用更新

这里是我的安装程序包列表

^{pr2}$

Answer 1

您已经在conda environment中安装了1.18.15 hdf5库，Tensorflow的某些依赖项有更新的hdf5 1.8.18头文件。在

尝试将您的condahdf5库更新到更新的1.8.18，或者使用以下命令：

set HDF5_DISABLE_VERSION_CHECK=2 

这将禁用警告，但不保证正确执行，因此最好将hdf5库更新到相同的版本1.8.18。在

Answer 2

我遇到了类似的错误消息，比如：

Headers are 1.10.3, library is 1.10.4

尝试通过运行以下命令在R中安装“rhd5”：

^{2}$

我通过在shell中设置环境变量“HDF5_DISABLE_VERSION_CHECK”来解决这个问题：

export HDF5_DISABLE_VERSION_CHECK=1

然后再次进入R，继续安装rhdf5。在