Correlation is not Causation, but it's a good start
The wonderful people at UPDB have opened source their full database. As such, I'm taking that data and running some analysis against it in order to enumerate potential correlational activity.
I use docker to download and dump the UPDB database. The database is large and it's not uploaded to this git repository so you will have to download it yourself.
final data will be located at ./out/uadb.csv
mkdir out && mkdir data
- Change the Makefile
dump
command by removing my mountpoint (C:\Users\faust...
) to your own. make build
make dump
It's all in the analyze.ipynb
Notebook.