Using the Keras RNN LSTM API for stock price prediction

Keras is a very easy-to-use high-level deep learning Python library running on top of other popular deep learning libraries, including TensorFlow, Theano, and CNTK. As you'll see soon, Keras makes building and playing with models a lot easier. To install and use Keras, along with TensorFlow as Keras' backend, it's best to set up a virtualenv first:

sudo pip install virtualenv

Then run the following commands, if you have TensorFlow 1.4 source on your machine and your iOS and Android apps; use the TensorFlow 1.4 custom library:

cd
mkdir ~/tf14_keras
virtualenv --system-site-packages ~/tf14_keras/
cd ~/tf14_keras/
source ./bin/activate
easy_install -U pip
pip install --upgrade https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-1.4.0-py2-none-any.whl
pip install keras

If you have TensorFlow 1.5 source on your machine, you should install TensorFlow 1.5 with Keras because the model created using Keras needs to have the same TensorFlow...