DwelltimeBootstrap¶

lumicks.pylake.population.dwelltime.DwelltimeBootstrap

class DwelltimeBootstrap(model: lumicks.pylake.population.dwelltime.DwelltimeModel, amplitude_distributions: numpy.ndarray, lifetime_distributions: numpy.ndarray)¶

Bootstrap distributions for a dwelltime model.

Important

This class should be initialized using lk.DwelltimeModel.calculate_bootstrap() and should not be constructed manually.

Warning

This is early access alpha functionality. While usable, this has not yet been tested in a large number of different scenarios. The API can still be subject to change without any prior deprecation notice! If you use this functionality keep a close eye on the changelog for any changes that may affect your analysis.

model¶

Original model sampled for the bootstrap distribution

Type: DwelltimeModel

amplitude_distributions¶

Array of sample optimized amplitude parameters; shape is [number of components, number of samples]

Type: np.ndarray

lifetime_distributions¶

Array of sample optimized lifetime parameters; shape is [number of components, number of samples]

Type: np.ndarray

Raises: ValueError – If the number of amplitude samples isn’t the same as the number of lifetime samples. If the number of parameters isn’t the same as the number of components.

calculate_stats(key, component, alpha=0.05)¶: Deprecated since version 0.13.3: This method will be removed in a future release. Use DwelltimeBootstrap.get_interval() to obtain the `1-alpha interval.

extend(iterations)¶

Extend the distribution by additional sampling iterations.

Parameters: iterations (int) – number of iterations (random samples) to add to the bootstrap distribution

get_interval(key, component, alpha=0.05)¶

Calculate the 1-alpha interval of the bootstrap distribution for a specified parameter.

NOTE: the 100*(1-alpha) % confidence intervals calculated here correspond to the alpha/2 and 1-(alpha/2) quantiles of the distribution. For distributions which are not well approximated by a normal distribution these values are not reliable confidence intervals.

Parameters

key ({'amplitude', 'lifetime'}) – name of the parameter to be analyzed
component (int) – index of the component to be analyzed
alpha (float) – confidence intervals are calculated as 100*(1-alpha)%

hist(*, n_bins=25, alpha=0.05, hist_kwargs=None, span_kwargs=None, line_kwargs=None)¶

Plot the bootstrap distributions for the parameters of a model.

Parameters

n_bins (int) – number of bins in the histogram
alpha (float) – confidence intervals are calculated as 100*(1-alpha)%
hist_kwargs (dict) – dictionary of plotting kwargs applied to histogram
span_kwargs (dict) – dictionary of plotting kwargs applied to the patch indicating the area spanned by the confidence intervals
line_kwargs (dict) – dictionary of plotting kwargs applied to the line indicating the distribution means

plot(alpha=0.05, n_bins=25, hist_kwargs=None, span_kwargs=None, line_kwargs=None)¶: Deprecated since version 0.13.3: This method has been renamed to more closely match its behavior. Use DwelltimeBootstrap.hist() instead.

property n_components¶: Number of components in the model.

property n_samples¶: Number of samples in the bootstrap.