A, gpu check
Check whether the installation cuda: nvcc -V
Then give up: the degree of your mother found CUDA8.0 does not support tensorflow 1.5, it is necessary to reduce version. (CUDA version also can improve, but it needs attention and cudnn CUDA version of correspondence, and is compatible with the computer's GPU, it will easily into the pit .cuda8 correspond cudnn6, cuda9 correspond cudnn7.)
Second, create a virtual environment:
conda create -n tf2 python = 3.6
Third, update pip
pip install pip -U # 升级 pip 到最新的版本 (>=10.0.0) 后进行配置
pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple //使用清华源
Fourth, the installation tensoflow2.0
python -m pip install --user --upgrade pip
pip install tensorflow==2.0.0-alpha0
执行import tensorflow报错：ModuleNotFoundError: No module named 'numpy.core._multiarray_umath'
解决：pip install --user --upgrade numpy
执行import tensorflow报错：TypeError: __init__() got an unexpected keyword argument 'serialized_options'