Metrics of Ensemble Forecast Skill (hydrostats.ens_metrics)

hydrostats.ens_metrics Module

Contains all the metrics included in hydrostats that measure forecast skill.

Each forecast metric is contained in a function, and every metric can treat missing values as well as remove zero and negative values from the timeseries data. Users will be warned which start dates have been removed in the warnings that display during the function execution.

Functions

auroc([fcst_ens, obs, threshold, ...])	Calculate Area Under the Relative Operating Characteristic curve (AUROC).
crps_hersbach(obs, fcst_ens[, remove_neg, ...])	Calculate the continuous ranked probability score (CRPS).
crps_kernel(obs, fcst_ens[, remove_neg, ...])	Compute the kernel representation of the continuous ranked probability score (CRPS).
ens_brier([fcst_ens, obs, threshold, ...])	Calculate the ensemble-adjusted Brier Score.
ens_crps(obs, fcst_ens[, adj, remove_neg, ...])	Calculate the ensemble-adjusted Continuous Ranked Probability Score (CRPS).
ens_mae(obs, fcst_ens[, remove_zero, ...])	Calculate the mean absolute error between observed values and the ensemble mean.
ens_me(obs, fcst_ens[, remove_zero, ...])	Calculate the mean error between observed values and the ensemble mean.
ens_mse(obs, fcst_ens[, remove_zero, ...])	Calculate the mean squared error between observed values and the ensemble mean.
ens_pearson_r(obs, fcst_ens[, remove_neg, ...])	Calculate the pearson correlation coefficient between observed values and the ensemble mean.
ens_rmse(obs, fcst_ens[, remove_zero, ...])	Calculate the root mean squared error between observed values and the ensemble mean.
skill_score(scores, bench_scores, perf_score)	Calculate the skill score of the given function.