How to Predict Stock Prices Easily - Intro to Deep Learning #7 by Siraj Raval on Youtube
This is the code for this video on Youtube by Siraj Raval part of the Udacity Deep Learning nanodegree. We use an LSTM neural network to predict the closing price of the S&P 500 using a dataset of past prices.
- keras
- tensorflow
Install Keras from here and Tensorflow from here.
Run this using jupyter notebook. Just type jupyter notebook in the main directory and the code will pop up in a browser window.
Use the price history AND two other metrics of your choice to predict the price of GOOGL stock with an LSTM network. You can find the CSV here. Metrics could be sentiment analysis from Twitter of what people have said about Google, dividends, etc.
Credits go to jaungiers. I've merely created a wrapper to get people started.
pip install --upgrade pip
pip install --upgrade tensorflow
pip install --upgrade keras
pip install --upgrade jupyter
jupyter notebook