elisa.plot.data#

Data classes for plotting.

class PlotData(name: str, result: FitResult, seed: int)[source]#

Bases: ABC

Base class for data used in plotting.

Attributes

ce_model

Point estimate of the folded source model.

back_counts
back_errors
back_ratio
ce_data
ce_errors
channel
channel_emax
channel_emean
channel_emid
channel_emin
channel_errors
channel_width
has_comps
ndata
net_counts
net_errors
params_dist
photon_egrid
spec_counts
spec_errors

Methods

`ce_model_ci`([cl])	Confidence/Credible intervals of the folded source model.
`pit`()	Probability integral transform.
`residuals`(rtype, seed, random_quantile, mle)	Residuals between the data and the fitted models.
`residuals_ci`(rtype, cl, seed, ...)	Confidence/Credible intervals of the residuals.
`residuals_sim`(rtype, seed, random_quantile)	Residuals bootstrap/ppc samples.

unfolded_model

property channel: ndarray#

property channel_emin: ndarray#

property channel_emax: ndarray#

property channel_emid: ndarray#

property channel_width: ndarray#

property channel_emean: ndarray#

property channel_errors: ndarray#

property photon_egrid: ndarray#

property spec_counts: ndarray | Array#

property spec_errors: ndarray | Array#

property back_ratio: float | ndarray | Array#

property back_counts: ndarray | Array | None#

property back_errors: ndarray | Array | None#

property net_counts: ndarray | Array#

property net_errors: ndarray | Array#

property ndata: int#

property ce_data: ndarray | Array#

property ce_errors: ndarray | Array#

abstract property ce_model: ndarray | Array#: Point estimate of the folded source model.

abstractmethod ce_model_ci(cl: float = 0.683) → ndarray | Array | None[source]#: Confidence/Credible intervals of the folded source model.

property has_comps: bool#

property params_dist: dict[str, ndarray | Array] | None#

abstractmethod unfolded_model(mtype: Literal['ne', 'ene', 'eene'], egrid: ndarray | Array, params: dict, comps: bool, cl: float | ndarray | Array | None = None) → ndarray | Array | dict[source]#

abstractmethod pit() → tuple[source]#: Probability integral transform.

abstractmethod residuals(rtype: Literal['rd', 'rp', 'rq'], seed: int | None, random_quantile: bool, mle: bool) → ndarray | Array | tuple[ndarray | Array, bool | ndarray | Array, bool | ndarray | Array][source]#: Residuals between the data and the fitted models.

abstractmethod residuals_sim(rtype: Literal['rd', 'rp', 'rq'], seed: int | None, random_quantile: bool) → ndarray | Array | None[source]#: Residuals bootstrap/ppc samples.

abstractmethod residuals_ci(rtype: Literal['rd', 'rp', 'rq'], cl: float, seed: int | None, random_quantile: bool, with_sign: bool) → ndarray | Array | None[source]#: Confidence/Credible intervals of the residuals.

class MLEPlotData(name: str, result: FitResult, seed: int)[source]#

Bases: PlotData

Attributes

`ce_model`	Point estimate of the folded source model.
`sign`	Sign of the difference between the data and the fitted models.

back_counts
back_errors
back_ratio
boot
ce_data
ce_errors
channel
channel_emax
channel_emean
channel_emid
channel_emin
channel_errors
channel_width
has_comps
ndata
net_counts
net_errors
params_dist
params_mle
photon_egrid
spec_counts
spec_errors

Methods

`ce_model_ci`([cl])	Confidence/Credible intervals of the folded source model.
`deviance`(rtype)	MLE and bootstrap deviance.
`model`(on_off, mtype)	Point estimate or bootstrap models of the on/off measurement.
`pit`()	Probability integral transform.
`residuals`(rtype[, seed, random_quantile, mle])	Residuals between the data and the fitted models.
`residuals_ci`(rtype[, cl, seed, ...])	Confidence/Credible intervals of the residuals.
`residuals_sim`(rtype[, seed, random_quantile])	Residuals bootstrap/ppc samples.

deviance_residuals_boot
deviance_residuals_mle
get_data_boot
get_model_boot
get_model_mle
pearson_residuals_boot
pearson_residuals_mle
quantile_residuals_mle
unfolded_model

result: MLEResult#

property boot: BootstrapResult#

property params_mle: dict[str, ndarray | Array]#

get_model_mle(name: str) → ndarray | Array[source]#

get_model_boot(name: str) → ndarray | Array | None[source]#

get_data_boot(name: str) → ndarray | Array | None[source]#

property ce_model: ndarray | Array#: Point estimate of the folded source model.

ce_model_ci(cl: float = 0.683) → ndarray | Array | None[source]#: Confidence/Credible intervals of the folded source model.

unfolded_model(mtype: Literal['ne', 'ene', 'eene'], egrid: ndarray | Array | None, params: dict | None, comps: bool, cl: float | ndarray | Array | None = None) → tuple[ndarray | Array | dict, ndarray | Array | dict | None][source]#

property sign: dict[str, ndarray | Array | None]#: Sign of the difference between the data and the fitted models.

model(on_off: Literal['on', 'off'], mtype: Literal['mle', 'boot']) → ndarray | Array | None[source]#: Point estimate or bootstrap models of the on/off measurement.

deviance(rtype: Literal['mle', 'boot']) → ndarray | Array | None[source]#: MLE and bootstrap deviance.

pit() → tuple[ndarray | Array, ndarray | Array][source]#: Probability integral transform.

residuals(rtype: Literal['rd', 'rp', 'rq'], seed: int | None = None, random_quantile: bool = True, mle: bool = True) → ndarray | Array | tuple[ndarray | Array, bool | ndarray | Array, bool | ndarray | Array][source]#: Residuals between the data and the fitted models.

residuals_sim(rtype: Literal['rd', 'rp', 'rq'], seed: int | None = None, random_quantile: bool = True) → ndarray | Array | None[source]#: Residuals bootstrap/ppc samples.

residuals_ci(rtype: Literal['rd', 'rp', 'rq'], cl: float = 0.683, seed: int | None = None, random_quantile: bool = True, with_sign: bool = False) → ndarray | Array | None[source]#: Confidence/Credible intervals of the residuals.

deviance_residuals_mle() → ndarray | Array[source]#

deviance_residuals_boot() → ndarray | Array | None[source]#

pearson_residuals_mle() → ndarray | Array[source]#

pearson_residuals_boot() → ndarray | Array | None[source]#

quantile_residuals_mle(seed: int, random: bool) → tuple[ndarray | Array, ndarray | Array | bool, ndarray | Array | bool][source]#

class PosteriorPlotData(name: str, result: FitResult, seed: int)[source]#

Bases: PlotData

Attributes

`ce_model`	Point estimate of the folded source model.
`sign`	Sign of the difference between the data and the fitted models.

back_counts
back_errors
back_ratio
ce_data
ce_errors
channel
channel_emax
channel_emean
channel_emid
channel_emin
channel_errors
channel_width
has_comps
ndata
net_counts
net_errors
params
params_dist
photon_egrid
ppc
spec_counts
spec_errors

Methods

`ce_model_ci`([cl])	Confidence/Credible intervals of the folded source model.
`deviance`(rtype)	Median, MLE, and ppc deviance.
`pit`()	Probability integral transform.
`residuals`(rtype[, seed, random_quantile, mle])	Residuals between the data and the fitted models.
`residuals_ci`(rtype[, cl, seed, ...])	Confidence/Credible intervals of the residuals.
`residuals_sim`(rtype[, seed, random_quantile])	Residuals bootstrap/ppc samples.

deviance_residuals_loo
deviance_residuals_median
deviance_residuals_mle
deviance_residuals_ppc
get_model_loo
get_model_median
get_model_mle
get_model_posterior
get_model_ppc
model
pearson_residuals_loo
pearson_residuals_median
pearson_residuals_mle
pearson_residuals_ppc
quantile_residuals
unfolded_model

result: PosteriorResult#

property params: dict[str, ndarray | Array]#

property ppc: PPCResult | None#

get_model_median(name: str) → ndarray | Array[source]#

get_model_loo(name: str) → ndarray | Array[source]#

get_model_posterior(name: str) → DataArray[source]#

get_model_ppc(name: str) → ndarray | Array | None[source]#

get_model_mle(name: str) → ndarray | Array | None[source]#

property ce_model: ndarray | Array#: Point estimate of the folded source model.

ce_model_ci(cl: float = 0.683) → ndarray | Array[source]#: Confidence/Credible intervals of the folded source model.

unfolded_model(mtype: Literal['ne', 'ene', 'eene'], egrid: ndarray | Array | None, params: dict | None, comps: bool, cl: float | ndarray | Array | None = None) → tuple[ndarray | Array | dict, ndarray | Array | dict | None][source]#

property sign: dict[str, ndarray | Array | None]#: Sign of the difference between the data and the fitted models.

model(on_off: Literal['on', 'off'], mtype: Literal['posterior', 'loo', 'median', 'mle', 'ppc']) → ndarray | Array | None[source]#

deviance(rtype: Literal['posterior', 'loo', 'mle', 'ppc']) → DataArray | None[source]#: Median, MLE, and ppc deviance.

pit() → tuple[source]#: Probability integral transform.

residuals(rtype: Literal['rd', 'rp', 'rq'], seed: int | None = None, random_quantile: bool = True, mle: bool = False) → ndarray | Array | tuple[ndarray | Array, bool | ndarray | Array, bool | ndarray | Array][source]#: Residuals between the data and the fitted models.

residuals_sim(rtype: Literal['rd', 'rp', 'rq'], seed: int | None = None, random_quantile: bool = True) → ndarray | Array | None[source]#: Residuals bootstrap/ppc samples.

residuals_ci(rtype: Literal['rd', 'rp', 'rq'], cl: float = 0.683, seed: int | None = None, random_quantile: bool = True, with_sign: bool = False) → ndarray | Array | None[source]#: Confidence/Credible intervals of the residuals.

deviance_residuals_loo() → ndarray | Array[source]#

deviance_residuals_median() → ndarray | Array[source]#

deviance_residuals_mle() → ndarray | Array[source]#

deviance_residuals_ppc() → ndarray | Array | None[source]#

pearson_residuals_loo() → ndarray | Array[source]#

pearson_residuals_median() → ndarray | Array[source]#

pearson_residuals_mle() → ndarray | Array[source]#

pearson_residuals_ppc() → ndarray | Array | None[source]#

quantile_residuals(seed: int, random: bool) → tuple[ndarray | Array, ndarray | Array | bool, ndarray | Array | bool][source]#