ravikg/keras-tensorflow-windows-installation

10 easy steps to install Tensorflow-GPU and Keras in Windows

MIT

Keras-TensorFlow-GPU-Windows-Installation

10 easy steps on the installation of TensorFlow-GPU and Keras in Windows 10 (Asus ROG GL502VS)

From tensorflow site: https://www.tensorflow.org/install/install_windows requirements for current(1.4.0) version :

CUDA Toolkit 8.0
cuDNN v6.0 Library
Python 2.7/3.5/3.6 (We will use 3.6)

Step 1: Install Anaconda Python 3.6 version

and follow the installation steps

Step 2: Update Anaconda

Open Anaconda Prompt (can be found in start menu) to type the following command(s)

conda update conda
conda update --all

Step 3: Install CUDA Tookit 8.0.

Choose your version depending on your Operating System.

Step 4: Download cuDNN.

Choose your version depending on your Operating System.
Membership registration is required.

Put your unzipped folder in C drive as follows:

C:\cudnn-8.0-windows10-x64-v6.0

Step 5: Add cuDNN into Environment PATH

Add the following path in your Environment.
Subjected to changes in your installation path.

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

Turn off all the prompts. Open a new Anaconda Prompt to type the following command(s)

echo %PATH%

You shall see that the new Environment PATH is there.

Step 6: Create an Anaconda environment with Python=3.6

Open Anaconda Prompt to type the following command(s)

conda create -n tensorflow python=3.6

Step 7: Activate the environment and install some common package

Open Anaconda Prompt to type the following command(s)

activate tensorflow
pip install numpy scipy matplotlib spyder jupyter pandas pillow scikit-learn h5py

Step 8: Install TensorFlow-GPU-1.4.0 (but the below command will install latest tensorflow)

Open Anaconda Prompt to type the following command(s)

pip install --ignore-installed --upgrade tensorflow-gpu

Step 9: Install Keras

Open Anaconda Prompt to type the following command(s)

pip install keras

Step 10: Testing

Let's try running ./examples/mnist_mlp.py

Open Anaconda Prompt to type the following command(s)

activate tensorflow
python mnist_mlp.py

Let's try running ./examples/mnist_cnn.py

Open Anaconda Prompt to type the following command(s)

activate tensorflow
python mnist_cnn.py

Congratulations ! You have successfully run Keras (with Tensorflow backend) over GPU on Windows !