HvsrSpatial

HvsrVault class definition.

class HvsrVault(coordinates, means, stddevs=None)

Bases: object

A container for Hvsr objects.

Variables

  • coordinates (ndarray) – Relative x and y coordinates of the sensors, where each row of the ndarray in an x, y pair.

  • means (ndarray) – Mean f0 value for each location, meaning is determined by distribution keyword argument.

  • stddevs (ndarray, optional) – Standard deviation for each location, meaning is determined by distribution keyword argument, default is None indicating no uncertainty is defined.

__init__(coordinates, means, stddevs=None)

Create a container for Hvsr statistics.

Parameters

  • coordinates (ndarray) – Relative x and y coordinates of the sensors, where each row of the ndarray in an x, y pair.

  • means (ndarray) – Mean f0 value for each location, meaning is determined by distribution keyword argument.

  • stddevs (ndarray, optional) – Standard deviation for each location, meaning is determined by distribution keyword argument, default is None indicating no uncertainty is defined.

  • distribution ({‘normal’, ‘lognormal’}, optional) – Distribution to which the mean and stddev for each point corresponds, default is ‘lognormal’.

bounded_voronoi(boundary)

Vertices of bounded Voronoi region.

Parameters

boundary (ndarray) – x, y coordinates defining the spatial boundary. Must be of shape (N, 2).

Returns

tuple – Of the form (new_vertices, indices) where new_vertices defines the vertices of each region and indices indicates how these vertices relate to the master statistics.

spatial_weights(boundary, dc_method='voronoi')

Calculate the weights for each Voronoi region.

Parameters

  • boundary (ndarray) – x, y coordinates defining the spatial boundary. Must be of shape (N, 2).

  • dc_method ({“voronoi”}, optional) – Declustering method, default is ‘voronoi’.

Returns

tuple – Of the form (weights, indices) where weights are the statistical weights and indicates the bounding box of each cell.

montecarlo_f0(mean, stddev, weights, dist_generators='lognormal', dist_spatial='lognormal', nrealizations=1000, generator='PCG64')

MonteCarlo simulation for spatial distribution of f0.

Parameters

  • mean, stddev (ndarray) – Mean and standard deviation of each generating point. Meaning of these parameters is dictated by dist_generators.

  • weights (ndarray) – Weights for each generating point.

  • dist_generators ({‘lognormal’, ‘normal’}, optional) – Assumed distribution of each generating point, default is lognormal.

    if dist is

    mean must be

    stddev must be

    normal

    \(\mu\)

    \(\sigma\)

    lognormal

    \(\lambda\)

    \(\zeta\)

  • dist_spatial ({‘lognormal’, ‘normal’}, optional) – Assumed distribution of spatial statistics on f0, default is lognormal.

  • generator ({‘PCG64’, ‘MT19937’}, optional) – Bit generator, default is PCG64.

Returns

tuple – Of the form (f0_mean, f0_stddev, f0_realizations).