Skip to content

Latest commit

 

History

History
executable file
·
13 lines (10 loc) · 530 Bytes

README.md

File metadata and controls

executable file
·
13 lines (10 loc) · 530 Bytes

Addition coopetition baseline

This repository will help you get started with coopetitions. The task is to modify the solution.py, so it stores the sum of two columns in a separate data.predict file into an output directory.

Metrics

Your solution will be evaluated by Mean Absolute Error metric:

from sklearn.metrics import mean_absolute_error
def scorer(solution, prediction):
    return mean_absolute_error(solution, prediction) 

It should give result as low as possible (absolute minimum is equal to zero).