numerai-reports 0.1.0

Reports for the Numerai machine learning competition

Numerai Reports

Collection of reports about the numer.ai machine learning competition.

This library was created to provide some reports and statistics about the competition, like round summaries and payout summaries, but also tools to analyze one's model performance. For now, all reports are pure numeric reports.

All information is retrieved via numerai's API and converted to pandas DataFrames, to make it easy to work with. This allows to create your own reports on top. numerai_reports also caches API results to disk to limit the amount of request and to speed-up report generation.

If you encounter a problem or have suggestions, feel free to open an issue.

Installation (TODO)

pip install --upgrade numerai_reports

Usage

Numerai_reports can be used as a regular, importable Python module or from the command line (TODO).

Python module

Usage example:

from numerai_reports import data, reports
lb = data.fetch_leaderboard(150, 159)
print(reports.payments(lb, "slyfox"))
print(reports.reputation(lb, "slyfox"))
print(reports.pass_rate(lb))
print(reports.out_of_n(lb))
print(reports.dominance("uuazed"))