Helper Module for Deep Learning.

Define common metrics.

class pynet.metrics.BinaryClassificationMetrics(score, thr=0.5, with_logit=True)[source]¶

Computes and stores the average and current value.

__init__(score, thr=0.5, with_logit=True)[source]¶

Initialize self. See help(type(self)) for accurate signature.

class pynet.metrics.SKMetrics(name, thr=0.5, with_logit=True, **kwargs)[source]¶

Wraping arounf scikit-learn metrics.

__init__(name, thr=0.5, with_logit=True, **kwargs)[source]¶

Initialize self. See help(type(self)) for accurate signature.

pynet.metrics.accuracy(y_pred, y)[source]¶

pynet.metrics.multiclass_dice(y_pred, y)[source]¶

Extension of the dice to a n classes problem.

pynet.metrics.pearson_correlation(y_pred, y)[source]¶

Pearson correlation.