Note

This page is a reference documentation. It only explains the function signature, and not how to use it. Please refer to the user guide for the big picture.

nidl.metrics.pearson_r

nidl.metrics.pearson_r(y_true, y_pred, sample_weight=None, multioutput='uniform_average', force_finite=False)[source]

Pearson correlation coefficient between 2 arrays y_true, y_pred. This score is symmetric between y_true and y_pred and is always between 1 (perfect correlation) and -1 (perfect anti-correlation).

Parameters:
y_true: array-like of shape (n_samples,) or (n_samples, n_outputs)

First input array.

y_pred: array-like of shape (n_samples,) or (n_samples, n_outputs)

Second input array.

sample_weightarray-like of shape (n_samples,), default=None

Sample weights for weighted Pearson’s correlation.

multioutput{‘raw_values’, ‘uniform_average’} or array-like of shape (n_outputs,), optional

Defines aggregating of multiple output scores.

  • ‘raw_values’: Returns a full set of scores in case of multioutput input.

  • ‘uniform_average’: Scores of all outputs are averaged with uniform weight.

  • array-like: Defines weights used to average scores.

force_finitebool, default=False

Flag indicating if NaN and -Inf scores resulting from constant data should be replaced with real numbers (1.0 if prediction is perfect, i.e. all data are equal , 0.0 otherwise). Default is False since Pearson’s correlation is not defined for constant data (zero variance).

Returns:
pearson_r: float or array of floats

The correlation score or ndarray of scores if ‘multioutput’ is ‘raw_values’.

Examples using nidl.metrics.pearson_r

Model probing callback of embedding estimators

Model probing callback of embedding estimators