Installation Environment
- Windows8.1
- python3.5.x (TensorFlow only supports version 3.5.x of Python on Windows)
- Pip 9.0.1
- tensorflow-gpu-0.12.0
- Cuda
Cudnn
installation Process
- Installing Python3.5
- Install Python
: https://www.python.org/
- Click on Windows under Downloads, this time using Python 3.5. Version 2
Select the fifth of these, Windows x86-64 executable installer download installation (it is best to add the environment variable when installing the hook, so that you do not need to configure the environment variables, follow the need for the place are ticked, all the way next)
2. Upgrade Pip
Open cmd input
python.exe-m pip Install--upgrade pip
(This use of the online installation, can also be downloaded to local, for local installation)
Installing the GPU version
Pip Install--upgrade https://Storage.googleapis.com/tensorflow/windows/gpu/tensorflow_ GPU-0.12.0-CP35-CP35M-WIN_AMD64.WHL
Installing the CPU version
Pip Install--upgrade https://storage.googleapis.com/tensorflow/windows/cpu/ TENSORFLOW-0.12.0-CP35-CP35M-WIN_AMD64.WHL
See if your GPU supports Cuda
GPU version to install the GPU support package (cuda8.0 address and cudnn5.1 address)
Installing cuda8.0
Install the corresponding CUDNN (this seems to need to login Nvdia account to be able to download, I was first registered)
Find the version you need, download the CUDNN compressed file, unzip it, add the extracted bin directory to the PATH environment variable.
$ python ... as tf>>> hello = tf.constant ('Hello, tensorflow! ' )>>> sess = TF. Session ()>>> print (Sess.run (hello)) Hello, TensorFlow!>>> a = tf.constant (10 )>>> b = tf.constant (+)>>> print (Sess.run (A + b)) the
TensorFlow installation process in Windows environment