How to Install Tensorflow on Windows10 with GPU support.


Links: 

  1. TensorFlow: https://www.tensorflow.org/versions/r0.12/get_started/os_setup.html#pip-installation-on-windows
  2. Anaconda 64-bit Python 3.5 https://www.continuum.io/downloads
  3. CUDA: https://developer.nvidia.com/cuda-downloads
  4. cuDNN: https://developer.nvidia.com/cudnn
Installation:
  1. Install CUDA v8
  2. Extract cuDNN folder and copy all files from bin folder to 'C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v8.0\bin', similarly from include to 'C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v8.0\include' and 'lib/x64' to  ''C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v8.0\lib/x64'.
  3. Install Anaconda, make sure you select Register Anaconda as system path for Python 3.5
  4. After installation is complete open command window as adminsitrator.
  5. Run 'pip install --upgrade virtualenv'
  6. Run 'virtualenv --system-site-packages c:\Tensorflow\tensorflowCPU0p12'
  7. Run 'c:\Tensorflow\tensorflowCPU0p12\Scripts\activate'
  8. Run 'pip install --upgrade https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow-0.12.0rc1-cp35-cp35m-win_amd64.whl'
  9. To test installation, enter command 'python' and then 'import tensorflow as tf'. It should run without error.

GPU Support:

  1. Run 'virtualenv --system-site-packages c:\Tensorflow\tensorflowGPU0p12'
  2. Run 'c:\Tensorflow\tensorflowGPU0p12\Scripts\activate'
  3. Run 'pip install --upgrade https://storage.googleapis.com/tensorflow/windows/gpu/tensorflow_gpu-0.12.0rc1-cp35-cp35m-win_amd64.whl'
  4. To test installation, enter command 'python' and then 'import tensorflow as tf'. It should run without error.


Comments

Popular posts from this blog

Build caffe, pycaffe and matcaffe on Windows

Running Tensorboard on Windows

Build TensorFlow on Windows