Codeforces is a well-known community for algorithms competitions. There are several heuristics that exist for finding suitable practice problems on Codeforces (sorting by the total number of solvers, for example), but none of them perform well and consistently in practice. We decided to create a better measure of problem difficulty, which we’ll call “problem rating”. You can read more about problem rating on our blog post here.

Scripts for a barebones version of individualized coding profiles can be found in the ui folder.

• ui/ contains scripts for creating individualized coding profiles.
• api_functions.py contains functions to parse information using APIs provided by codeforces.
• elo.py was used to calculate problem ratings.
• blogPost_v1.ipynb is a jupyter/iPython notebook that contains Python and R scripts used to generate figures for our first blog post.
• prob_duplicates.py identifies probable duplicated problems.

• problem_duplicates.csv A list of duplicated problems between Div. 1 and Div. 2.
• problem_ratings.csv Problem ratings from elo.py.
• rating_histories.csv Rating history of active users who competed in at least 15 competitions.