/RaphaelGervillie-graph-forecast-national-elections

2017 French Presidential Elections Forecast : France, like many countries, operates by the system of universal suffrage. During each vote, the results are final when all the votes have finished being counted. But some estimation can be made on the final results during the day. This challenge proposes to explore models allowing to predict the results using only half of the votes. This type of challenge is part of all the economic issues that are seeing a revival with the arrival of machine learning methods. With the increase of available data in open access, it is possible to better model complex economic problems. This challenge therefore aims on the one hand to improve the estimate of the election results and on the other hand to propose an original model allowing to capture the multiple dimensions of this problem.

Primary LanguageJupyter Notebook

Graph-based forecast of French National Elections 2017 results (second tour)

Authors : Raphaël Gervillié and Gaspard Michel

Institut Polytechnique de Paris - M2 Data Science

alt text

alt text

Getting started

Install

To run a submission and the notebook you will need the dependencies listed in requirements.txt. You can install install the dependencies with the following command-line:

pip install -U -r requirements.txt

Challenge description

Get started with the dedicated notebook

Test a submission

The submissions need to be located in the submissions folder. For instance for my_submission, it should be located in submissions/my_submission.

To run a specific submission, you can use the ramp-test command line:

ramp-test --submission my_submission

You can get more information regarding this command line:

ramp-test --help

To go further

You can find more information regarding ramp-workflow in the dedicated documentation