Every year in France, more than 800 000 students aged around 15 take a compulsory national exam, the "Brevet des collèges" which marks the end of four years of studies. The national average is 86.5% of success. This project aims at predicting the average success rate at the brevet des colleges for each french public college and identifying factors of the success of the best colleges and the degrees of flexibility of the government to reduce inequalities.
Open a terminal and
- install the ramp-workflow library (if not already done)
pip install git+https://github.com/paris-saclay-cds/ramp-workflow.git
- Follow the ramp-kits instructions from the wiki
-
Get started on this RAMP with the starting kit notebook
-
The notebook data_camp_get_data gives the scrapping code for the target.
-
WARNING: The maps are not visible in the notebook on the github because they are dynamic. So you need to run the notebook locally and make sure you have an internet connection because they need to access some url on FranceGeoJson where some dynamic maps are held. To see the notebook correctly compiled with the dynamics maps, please click here.
-
NOTE: In order to properly run the data_camp_get_data.ipynb notebook, please download the following file: https://drive.google.com/open?id=1kquU3w7GADjwEqP7rbcqRwBtHeY2HO-Z and place it in ./data/donnees_geographiques/donnees_postal_code Data source: https://data.opendatasoft.com/explore/dataset/code-postal-code-insee-2015%40public/table/?fbclid=IwAR2n_ZKM09d5GAFOcu5vFpadj-yOR3kpmFIqgiEgskY7YvYvNMGgq-Pt1CQ
-
To test the starting-kit, please run the following command
ramp_test_submission --starting_kit