Uncertainty Metric

class stonesoup.metricgenerator.uncertaintymetric.SumofCovarianceNormsMetric(tracks_key: str = 'tracks', generator_name: str = 'sumCovNorms_generator')[source]

Bases: _CovarianceNormsMetric

Computes the sum of the covariance matrix norms of each state at a time step. The matrix norm calculated is the Frobenius norm. The metric generator will return this value at each time step in the track(s) as a measure of the uncertainty.

Parameters:
  • tracks_key (str, optional) – Key to access set of tracks added to MetricManager

  • generator_name (str, optional) – Unique identifier to use when accessing generated metrics from MultiManager

generator_name: str

Unique identifier to use when accessing generated metrics from MultiManager

compute_covariancenorms(track_states)[source]

Computes the sum of covariance norms metric for a single time step.

Parameters:

track_states (list of State) – List of states created by a filter

Returns:

metric – The sum of covariance matrix norms metric at a single time step

Return type:

SingleTimeMetric

class stonesoup.metricgenerator.uncertaintymetric.MeanofCovarianceNormsMetric(tracks_key: str = 'tracks', generator_name: str = 'covNorms_generator')[source]

Bases: _CovarianceNormsMetric

Parameters:
  • tracks_key (str, optional) – Key to access set of tracks added to MetricManager

  • generator_name (str, optional) – Unique identifier to use when accessing generated metrics from MultiManager

compute_covariancenorms(track_states)[source]

Computes the mean of covariance norms metric for a single time step.

Parameters:

track_states (list of State) – List of states created by a filter

Returns:

metric – The mean of covariance matrix norms metric at a single time step

Return type:

SingleTimeMetric