We can tie together two charts through a trend detections score that we call ETA.
To start, we use data in this format:
date,count,eta 2014-08-24 00:00:00,76.0,0 2014-08-24 02:00:00,91.0,0.34 2014-08-24 04:00:00,69.0,0.45 2014-08-24 06:00:00,76.0,0.17 2014-08-24 08:00:00,39.0,0.84 2014-08-24 10:00:00,36.0,0.094 2014-08-24 12:00:00,59.0,0.77 2014-08-24 14:00:00,58.0,0.025 2014-08-24 16:00:00,58.0,0
The config file is available in js/app.js, which is where we indicate the data path:
var data_path = './data/scotus_analyzed.csv';
To run the chart on a local server, we can visit the tye_chart directory and run a server.
$cd trend_detection_graph/src/tye_chart $python -m SimpleHTTPServer 8080 ...OR... $python3 -m http.server 8080
We then open a browser to localhost:8080.
To use anacaonda to manage this enviroment, we can build tye_chart_env
from the environment.yml file saved in this repo:
$ conda env create -f environment.yml $ source activate tye_chart_env
A .pth file must be placed here:
~/anaconda/envs/tye_chart_env/lib/python3.5/site-packages
This file described below allows python to locate modules without needing to set $PYTHONPATH each time.
$ cat tye_chart_env.pth <local path>/trend_detection_graph/src <local path>/trend_detection_graph/src/ipython_module