MVHistogram (class)
Treatment under-the-hood:
-
When sampling, a random value between 0 and 1 is drawn. The N-dimensional bins are then unraveled and integrated to create a flattened cdf. The cdf is then linearly interpolated to find the index of the unraveled bins in which to sample, as well as the relative location in the bin. The selected bin is then artificially subdivided by the same shape grid as the original binning and linearly interpolated based on the remainder to return a single value for MVHistogram.sample.
-
Means and covariances (see MVHistogram.calculate_means_covariances, MVHistogram.calculate_means, MVHistogram.calculate_covariances) are calculated by sampling (with a default size of 1e5), and determining the mean and covariances on that sample.
- arccos
- arcsin
- arctan
- arctan2
- bins
- cached_sample
- calculate_covariances
- calculate_means
- calculate_means_covariances
- cdf
- clear_cached_sample
- copy
- cos
- deepcopy
- density
- dist_constructor_args
- dist_constructor_func
- dist_constructor_object
- from_data
- get_wrap_at
- labels
- labels_latex
- log
- log10
- logcdf
- logpdf
- ndimensions
- plot
- plot_cdf
- plot_gaussian
- plot_pdf
- plot_sample
- plot_uncertainties
- ppf
- sample
- sin
- slice
- take_dimensions
- tan
- to_dict
- to_file
- to_gaussian
- to_histogram
- to_json
- to_mvgaussian
- to_mvsamples
- to_samples
- to_univariate
- uncertainties
- uniqueid
- units
- wrap_ats