Michael Boker's capstone project for the Udacity Machine Learning Engineer Nanodegree

In this project, I create a recurrent neural network used for predicting behavior of the Dow Jones Industrial Average on a day, based on headlines in the news.

This project requires an anaconda environment with the following packages and their dependencies: numpy 1.12.1 pandas 0.19.2 matplotlib 2.0.0 tensorflow 1.1.0 scikit-learn 0.18.2 tqdm 4.11.2 nltk 3.2.2 **nltk requires installation of languages. To accomplish this, after installing nltk in your environment using pip, run python in the terminal. Through the python REPL, import nltk, then run nltk.install. Through the GUI, install all languages and corpuses.

The final report for this project is in the documents directory, titled Capstone_Report.pdf All code is contained in the code directory. "Capstone Notebook.ipynb" shows how the code was used, step by step, with the exception of visualization code. Most of the visualization code was run through the REPL.