Install Tensorflow-1.4 on Windows with GPU

[jinzishuai]

https://www.tensorflow.org/install/install_windows

• [x] CUDA® Toolkit 8.0. For details, see NVIDIA's documentation Ensure that you append the relevant Cuda pathnames to the %PATH% environment variable as described in the NVIDIA documentation.

• [x] The NVIDIA drivers associated with CUDA Toolkit 8.0.

• [x] cuDNN v6.1. For details, see NVIDIA's documentation. Note that cuDNN is typically installed in a different location from the other CUDA DLLs. Ensure that you add the directory where you installed the cuDNN DLL to your %PATH% environment variable.

• [x] Install Python-3.6

• [x] install tensorflow with GPU: C:\Users\ShiJin>pip3 install --upgrade tensorflow-gpu

[jinzishuai]

1. CUDA-8

http://docs.nvidia.com/cuda/cuda-installation-guide-microsoft-windows/

• Download & install CUDA-8.0-GA2 from https://developer.nvidia.com/cuda-80-ga2-download-archive
  • install base
  • install patch2

[jinzishuai]

Warning without Visual Studio

[jinzishuai]

2. cuDNN 6 for CUDA-8 and Windows 10

https://developer.nvidia.com/compute/machine-learning/cudnn/secure/v6/prod/8.0_20170307/cudnn-8.0-windows10-x64-v6.0-zip

Extracted as E:\cudnn-8.0-windows10-x64-v6.0\cuda

Add to System Path

E:\cudnn-8.0-windows10-x64-v6.0\cuda\bin

[jinzishuai]

3. Install Python 3.6

https://www.python.org/downloads/release/python-363/

Choose to disable the max length limit

[jinzishuai]

Review of Paths

[jinzishuai]

[jinzishuai]

Nvidia Diver Mode

On Windows 7 and later, the operating system provides two driver models under which the NVIDIA Driver may operate: ‣ The WDDM driver model is used for display devices. ‣ The Tesla Compute Cluster (TCC) mode of the NVIDIA Driver is available for nondisplay devices such as NVIDIA Tesla GPUs, and the GeForce GTX Titan GPUs; it uses the Windows WDM driver model.