• running directly on a bokeh-server
• embedded into a separate Flask application

You will need to first download some sample data, then follow the instructions for running the example.

Sample Data

To run the stocks applet example, you first need to download a sample data file. The file is located at:

http://quantquote.com/files/quantquote_daily_sp500_83986.zip

A python script is included in the applet's directory to collect and extract the data, which can be run directly from the directory:

python stock_data.py

You can use a browser to download the file, or depending on your system you may be able to use curl:

curl -O http://quantquote.com/files/quantquote_daily_sp500_83986.zip

or wget:

wget http://quantquote.com/files/quantquote_daily_sp500_83986.zip

from the command line. Once the file is downloaded you should move it to this directory, and unzip it. You may be able to unzip the file by clicking on it, or by executing this command from the command line:

unzip quantquote_daily_sp500_83986.zip

This should leave a "quantquote_daily_sp500_83986" subdirectory in this directory. Move the 'daily' directory by executing this command from the command line:

mv quantquote_daily_sp500_83986/daily .

Now you can safely remove the empty "quantquote_daily_sp500_83986" directory.

Running

Bokeh Server

To view this applet directly from a bokeh server, you simply need to run a bokeh-server and point it at the stock example script:

bokeh-server --script stock_app.py

Now navigate to the following URL in a browser:

http://localhost:5006/bokeh/stocks

Flask Application

To embed this applet into a Flask application, first you need to run a bokeh-server and point it at the stock example script. In this directory, execute the command:

bokeh-server --script stock_app.py

Next you need to run the flask server that embeds the stock applet:

python flask_server.py

Now you can see the stock correlation applet by navigating to the following URL in a browser:

http://localhost:5050/

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| the ui is a fork from bokeh-server example .

extract the app in your home directory

for best result have the following packages running in your unix environment -anconda distribution python 2.7(should have everything) -bokeh-server may need to be added

have not trained the full list of fortune five hundred companies

only a few have been tarined and predicted . feel free to train the reset using the algorithm in the ipython notebook testing stuff .

to train the next set of stock modify the lists in stock_app.py and neural_net_train&store_data.py execute neural_net_train&store_data.py to predict and sotre reuslts automaticly

have fun and feel free to extend it to your needs

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||