Competition evaluation metric for Python

In case anyone is interested. Feel free to suggest improvements or catch anything we missed. Using this code for model validation. #..written for python 2.7 import numpy as np import pandas as pd def CRPS_row(row): “”” This function is purpose-built for Kaggle. With a couple tweaks it can be generalized for other applications. row should be a 601-element list where the first element is the true volme and the subsequent elements are cumulatively summed probabilities. “”” V_m = row[0] p = np.array(row[1:]) v = np.array(range(len(n))) h = v >= V_m sq_dists = (p – h)**2 return(np.sum(sq_dists)/len(sq_dists)) def CRPS_mean(df): “”” Function recieves pandas dataframe as an input with the first column being a column of truths (V_m) and the subsequent 600 columns being cumulatively summed probabilities “”” crps_vec = df.apply(CRPS_row, axis…


Link to Full Article: Competition evaluation metric for Python