skill_score¶
-
hydrostats.ens_metrics.
skill_score
(scores, bench_scores, perf_score, eff_sample_size=None, remove_nan_inf=False)¶ Calculate the skill score of the given function.
Parameters: - scores (float or ndarray) – The verification scores, or the mean of the verification scores in an ndarray (float).
- bench_scores (float or ndarray) – The reference or benchmark verification scores, or the mean of the benchmark scores (float).
- perf_score (int or float) – The perfect score of the score, typically 1 or 0.
- eff_sample_size (float) – The effective sample size of the data to be used when estimating the sampling uncertainty. Default is None, which will set the eff_sample_size to the length of scores.
- remove_nan_inf (bool) – If True, removes NaN and Inf values in the scores if they exist, pairwise. If False (default), the function will raise an exception.
Returns: Dictionary containing: {“skillScore”: Float, the skill score, “standardDeviation”: Float, the estimated standard deviation of the skill score} If the scores and bench scores given were floats, the standard deviation will be NaN.
Return type: dict
References
- Stefan Siegert (2017). SpecsVerification: Forecast Verification Routines for Ensemble Forecasts of Weather and Climate. R package version 0.5-2. https://CRAN.R-project.org/package=SpecsVerification
Examples