TensorFlow测试程序报异常：FutureWarning: Conversion of the second argument of issubdtype from `float` to `np

使用安装好的tensorflow-gpu 进行程序测试时出现异常:

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`.

  from ._conv import register_converters as _register_converters

原因:

    numpy包的版本和h5py包版本不匹配导致

    我的本地numpy版本为1.14.1，h5py版本为2.7.1

解决方案:

方法一: 降级numpy版本

    pip install numpy==1.13.0

方法二: 升级h5py版本

    pip install h5py==2.8.0rc1