Covariance

class gammapy.modeling.Covariance(parameters, data=None)

Bases: object

Parameter covariance class.

Parameters:

parametersParameters

Parameter list.

datandarray

Covariance data array.

Attributes Summary

correlation	Correlation matrix as a numpy.ndarray.
data	Covariance data as a ndarray.
scipy_mvn
shape	Covariance shape.

Methods Summary

from_factor_matrix(parameters, matrix)	Set covariance from factor covariance matrix.
from_stack(covar_list)	Stack sub-covariance matrices from list.
get_subcovariance(parameters)	Get sub-covariance matrix.
plot_correlation([figsize])	Plot correlation matrix.
set_subcovariance(covar)	Set sub-covariance matrix.

Attributes Documentation

correlation

Correlation matrix as a numpy.ndarray.

Correlation \(C\) is related to covariance \(\Sigma\) via:

\[C_{ij} = \frac{ \Sigma_{ij} }{ \sqrt{\Sigma_{ii} \Sigma_{jj}} }\]

data

Covariance data as a ndarray.

scipy_mvn

shape

Covariance shape.

Methods Documentation

classmethod from_factor_matrix(parameters, matrix)

Set covariance from factor covariance matrix.

Used in the optimizer interface.

classmethod from_stack(covar_list)

Stack sub-covariance matrices from list.

Parameters:

covar_listlist of Covariance

List of sub-covariances.

Returns:

covarCovariance

Stacked covariance.

get_subcovariance(parameters)

Get sub-covariance matrix.

Parameters:

parametersParameters

Sub list of parameters.

Returns:

covariancendarray

Sub-covariance.

plot_correlation(figsize=None, **kwargs)

Plot correlation matrix.

Parameters:

figsizetuple, optional

Figure size. Default is None, which takes (number_params*0.9, number_params*0.7).

**kwargsdict

Keyword arguments passed to plot_heatmap.

Returns:

axAxes, optional

Matplotlib axes.

set_subcovariance(covar)

Set sub-covariance matrix.

Parameters:

covarCovariance

Sub-covariance.