yasa.EpochByEpochAgreement.get_agreement#

EpochByEpochAgreement.get_agreement(sample_weight=None, scorers=None, pooled=False)[source]#

Return a pandas.DataFrame of weighted (i.e., averaged) agreement scores.

Parameters:

sample_weightNone or pandas.Series: Sample weights passed to underlying sklearn.metrics functions where possible. If a pandas.Series, the index must match exactly that of data, which excludes ART and UNS epochs.
scorersNone, list, or dictionary: The scorers to be used for evaluating agreement. If None (default), default scorers are used. If a list of strings, each name is mapped to the sklearn.metrics.<name>_score function (e.g. "accuracy", "cohen_kappa"), called with sample_weight; metrics that require extra arguments (e.g. average for precision) must be passed as a dictionary instead. If a dictionary, keys are scorer names (str) and values are functions taking 3 positional arguments (true values, predicted values, and sample weights).
pooledbool: If False (default), agreement scores are computed per session and returned as a DataFrame with one row per session. If True, all epochs across all sessions are pooled before computing a single set of agreement scores, returned as a Series.

Returns:

agreementpandas.DataFrame or pandas.Series

If pooled=False, a DataFrame with agreement metrics as columns and sessions as rows. If pooled=True, a Series with agreement metrics as index.

With the default scorers, the proportion-based metrics (accuracy, balanced_acc, precision, f1) are expressed as percentages (0-100). kappa and mcc are correlation-like coefficients ranging from -1 to 1. Custom scorers are returned as is.