MaximumMeanDiscrepancy#
- class bayesflow.metrics.MaximumMeanDiscrepancy(name: str = 'maximum_mean_discrepancy', kernel: str = 'inverse_multiquadratic', unbiased: bool = False, **kwargs)[source]#
Bases:
Metric- result()[source]#
Compute the current metric value.
- Returns:
A scalar tensor, or a dictionary of scalar tensors.
- __call__(*args, **kwargs)#
Call self as a function.
- add_variable(shape, initializer, dtype=None, aggregation='sum', name=None)#
- add_weight(shape=(), initializer=None, dtype=None, name=None)#
- property dtype#
- classmethod from_config(config)#
- get_config()#
Return the serializable config of the metric.
- reset_state()#
Reset all of the metric state variables.
This function is called between epochs/steps, when a metric is evaluated during training.
- stateless_reset_state()#
- stateless_result(metric_variables)#
- stateless_update_state(metric_variables, *args, **kwargs)#
- property variables#