SpectrumDataset

class gammapy.spectrum.SpectrumDataset(model=None, counts=None, livetime=None, aeff=None, edisp=None, background=None, mask_safe=None, mask_fit=None, name='', gti=None)

Bases: gammapy.modeling.Dataset

Spectrum dataset for likelihood fitting.

The spectrum dataset bundles reduced counts data, with a spectral model, background model and instrument response function to compute the fit-statistic given the current model and data.

Parameters:

model : SpectralModel

Fit model

counts : CountsSpectrum

Counts spectrum

livetime : Quantity

Livetime

aeff : EffectiveAreaTable

Effective area

edisp : EnergyDispersion

Energy dispersion

background : CountsSpectrum

Background to use for the fit.

mask_safe : ndarray

Mask defining the safe data range.

mask_fit : ndarray

Mask to apply to the likelihood for fitting.

name : str

Dataset name.

gti : GTI

GTI of the observation or union of GTI if it is a stacked observation

See also

SpectrumDatasetOnOff, FluxPointsDataset, gammapy.cube.MapDataset

Attributes Summary

data_shape	Shape of the counts data
energy_range	Energy range defined by the safe mask
excess	Excess (counts - alpha * counts_off)
likelihood_type
mask	Combined fit and safe mask
mask_safe
model
parameters

Methods Summary

copy(self)	A deep copy.
create(e_reco[, e_true, reference_time])	Creates empty SpectrumDataset
fake(self[, random_state])	Simulate fake counts for the current model and reduced irfs.
likelihood(self)	Total likelihood given the current model parameters.
likelihood_per_bin(self)	Likelihood per bin given the current model parameters
npred(self)	Return npred map (model + background)
plot_counts(self[, ax])	Plot predicted and detected counts.
plot_fit(self)	Plot counts and residuals in two panels.
plot_residuals(self[, method, ax])	Plot residuals.
residuals(self[, method])	Compute the spectral residuals.
stack(self, other)	Stack this dataset with another one.

Attributes Documentation

data_shape

Shape of the counts data

energy_range

Energy range defined by the safe mask

excess

Excess (counts - alpha * counts_off)

likelihood_type = 'cash'

mask

Combined fit and safe mask

mask_safe

model

parameters

Methods Documentation

copy(self)

A deep copy.

classmethod create(e_reco, e_true=None, reference_time='2000-01-01')

Creates empty SpectrumDataset

Empty containers are created with the correct geometry. counts, background and aeff are zero and edisp is diagonal.

The safe_mask is set to False in every bin.

Parameters:

e_reco : Quantity

edges of counts vector

e_true : Quantity

edges of effective area table. If not set use reco energy values. Default : None

reference_time : Time

reference time of the dataset, Default is “2000-01-01”

fake(self, random_state='random-seed')

Simulate fake counts for the current model and reduced irfs.

This method overwrites the counts defined on the dataset object.

Parameters:

random_state : {int, ‘random-seed’, ‘global-rng’, RandomState}

Defines random number generator initialisation. Passed to get_random_state.

likelihood(self)

Total likelihood given the current model parameters.

likelihood_per_bin(self)

Likelihood per bin given the current model parameters

npred(self)

Return npred map (model + background)

plot_counts(self, ax=None)

Plot predicted and detected counts.

Parameters:

ax : Axes

Axes object.

Returns:

ax : Axes

Axes object.

plot_fit(self)

Plot counts and residuals in two panels.

Calls plot_counts and plot_residuals.

plot_residuals(self, method='diff', ax=None, **kwargs)

Plot residuals.

Parameters:

ax : Axes

Axes object.

method : {“diff”, “diff/model”, “diff/sqrt(model)”}

Normalization used to compute the residuals, see SpectrumDataset.residuals()

**kwargs : dict

Keywords passed to CountsSpectrum.plot()

Returns:

ax : Axes

Axes object.

residuals(self, method='diff')

Compute the spectral residuals.

Parameters:

method : {“diff”, “diff/model”, “diff/sqrt(model)”}

Method used to compute the residuals. Available options are:

• diff (default): data - model
• diff/model: (data - model) / model
• diff/sqrt(model): (data - model) / sqrt(model)

Returns:

residuals : CountsSpectrum

Residual spectrum

stack(self, other)

Stack this dataset with another one.

Safe mask is applied to compute the stacked counts vector. Counts outside each dataset safe mask are lost.

Stacking is performed in-place.

The stacking of 2 datasets is implemented as follows. Here, \(k\) denotes a bin in reconstructed energy and \(j = {1,2}\) is the dataset number

The mask_safe of each dataset is defined as:

\[\begin{split}\epsilon_{jk} =\left\{\begin{array}{cl} 1, & \mbox{if bin k is inside the energy thresholds}\\ 0, & \mbox{otherwise} \end{array}\right.\end{split}\]

Then the total counts and model background bkg are computed according to:

\[ \begin{align}\begin{aligned}\overline{\mathrm{n_{on}}}_k = \mathrm{n_{on}}_{1k} \cdot \epsilon_{1k} + \mathrm{n_{on}}_{2k} \cdot \epsilon_{2k}\\\overline{bkg}_k = bkg_{1k} \cdot \epsilon_{1k} + bkg_{2k} \cdot \epsilon_{2k}\end{aligned}\end{align} \]

The stacked safe_mask is then:

\[\overline{\epsilon_k} = \epsilon_{1k} OR \epsilon_{2k}\]

Please refer to the IRFStacker for the description of how the IRFs are stacked.

Parameters:

other : SpectrumDataset

the dataset to stack to the current one