threeML.utils.time_series package

Submodules

threeML.utils.time_series.binned_spectrum_series module

class threeML.utils.time_series.binned_spectrum_series.BinnedSpectrumSeries(binned_spectrum_set, first_channel=1, ra=None, dec=None, mission=None, instrument=None, verbose=True)

Bases: threeML.utils.time_series.time_series.TimeSeries

property binned_spectrum_set

returns the spectrum set :return: binned_spectrum_set

property bins

the time bins of the spectrum set :return: TimeIntervalSet

count_per_channel_over_interval(start, stop)

return the number of counts in the selected interval :param start: start of interval :param stop: stop of interval :return:

counts_over_interval(start, stop)

return the number of counts in the selected interval :param start: start of interval :param stop: stop of interval :return:

exposure_over_interval(start, stop)

calculate the exposure over the given interval

Parameters

• start – start time

• stop – stop time

Returns

set_active_time_intervals(*args)

Set the time interval(s) to be used during the analysis. Specified as ‘tmin-tmax’. Intervals are in seconds. Example:

set_active_time_intervals(“0.0-10.0”)

which will set the time range 0-10. seconds.

view_lightcurve(start: float = - 10, stop: float = 20.0, dt: float = 1.0, use_binner: bool = False) → None

Parameters

• start –

• stop –

• dt –

• use_binner –

threeML.utils.time_series.event_list module

class threeML.utils.time_series.event_list.EventList(arrival_times, measurement, n_channels, start_time=None, stop_time=None, native_quality=None, first_channel=0, ra=None, dec=None, mission=None, instrument=None, verbose=True, edges=None)

Bases: threeML.utils.time_series.time_series.TimeSeries

property arrival_times

bin_by_bayesian_blocks(start, stop, p0, use_background=False)

bin_by_constant(start, stop, dt=1)

Interface to the temporal binner’s constant binning mode

Parameters

• start – start time of the bins

• stop – stop time of the bins

• dt – temporal spacing of the bins

Returns

bin_by_custom(start, stop)

Interface to temporal binner’s custom bin mode

Parameters

• start – start times of the bins

• stop – stop times of the bins

Returns

bin_by_significance(start, stop, sigma, mask=None, min_counts=1)

Interface to the temporal binner’s significance binning model

param start

start of the interval to bin on

param stop

stop of the interval ot bin on

param sigma

sigma-level of the bins

param mask

(bool) use the energy mask to decide on ,significance

param min_counts

minimum number of counts per bin

return

property bins

count_per_channel_over_interval(start, stop)

Parameters

• start –

• stop –

Returns

counts_over_interval(start, stop)

return the number of counts in the selected interval :param start: start of interval :param stop: stop of interval :return:

property measurement

property n_events

view_lightcurve(start: float = - 10, stop: float = 20.0, dt: float = 1.0, use_binner: bool = False) → None

Parameters

• start –

• stop –

• dt –

• use_binner –

class threeML.utils.time_series.event_list.EventListWithDeadTime(arrival_times, measurement, n_channels, start_time=None, stop_time=None, dead_time=None, first_channel=0, quality=None, ra=None, dec=None, mission=None, instrument=None, verbose=True, edges=None)

Bases: threeML.utils.time_series.event_list.EventList

exposure_over_interval(start, stop)

calculate the exposure over the given interval

Parameters

• start – start time

• stop – stop time

Returns

set_active_time_intervals(*args)

Set the time interval(s) to be used during the analysis.

Specified as ‘tmin-tmax’. Intervals are in seconds. Example:

set_active_time_intervals(“0.0-10.0”)

which will set the energy range 0-10. seconds.

class threeML.utils.time_series.event_list.EventListWithDeadTimeFraction(arrival_times, measurement, n_channels, start_time=None, stop_time=None, dead_time_fraction=None, first_channel=0, quality=None, ra=None, dec=None, mission=None, instrument=None, verbose=True, edges=None)

Bases: threeML.utils.time_series.event_list.EventList

exposure_over_interval(start, stop)

calculate the exposure over the given interval

Parameters

• start – start time

• stop – stop time

Returns

set_active_time_intervals(*args)

Set the time interval(s) to be used during the analysis.

Specified as ‘tmin-tmax’. Intervals are in seconds. Example:

set_active_time_intervals(“0.0-10.0”)

which will set the energy range 0-10. seconds.

class threeML.utils.time_series.event_list.EventListWithLiveTime(arrival_times, measurement, n_channels, live_time, live_time_starts, live_time_stops, start_time=None, stop_time=None, quality=None, first_channel=0, rsp_file=None, ra=None, dec=None, mission=None, instrument=None, verbose=True, edges=None)

Bases: threeML.utils.time_series.event_list.EventList

exposure_over_interval(start, stop)

Parameters

• start – start time of interval

• stop – stop time of interval

Returns

exposure

set_active_time_intervals(*args)

Set the time interval(s) to be used during the analysis.

Specified as ‘tmin-tmax’. Intervals are in seconds. Example:

set_active_time_intervals(“0.0-10.0”)

which will set the energy range 0-10. seconds.

exception threeML.utils.time_series.event_list.OverLappingIntervals

Bases: RuntimeError

exception threeML.utils.time_series.event_list.ReducingNumberOfSteps

Bases: Warning

exception threeML.utils.time_series.event_list.ReducingNumberOfThreads

Bases: Warning

threeML.utils.time_series.event_list.ceildiv(a, b)

threeML.utils.time_series.polynomial module

exception threeML.utils.time_series.polynomial.CannotComputeCovariance

Bases: RuntimeWarning

class threeML.utils.time_series.polynomial.PolyBinnedLogLikelihood(x, y, model, exposure)

Bases: threeML.utils.time_series.polynomial.PolyLogLikelihood

Implements a Poisson likelihood (i.e., the Cash statistic). Mind that this is not the Castor statistic (Cstat). The difference between the two is a constant given a dataset. I kept Cash instead of Castor to make easier the comparison with ROOT during tests, since ROOT implements the Cash statistic.

class threeML.utils.time_series.polynomial.PolyLogLikelihood(model, exposure)

Bases: object

class threeML.utils.time_series.polynomial.PolyUnbinnedLogLikelihood(events, model, t_start, t_stop, exposure)

Bases: threeML.utils.time_series.polynomial.PolyLogLikelihood

Implements a Poisson likelihood (i.e., the Cash statistic). Mind that this is not the Castor statistic (Cstat). The difference between the two is a constant given a dataset. I kept Cash instead of Castor to make easier the comparison with ROOT during tests, since ROOT implements the Cash statistic.

class threeML.utils.time_series.polynomial.Polynomial(coefficients, is_integral=False)

Bases: object

property coefficients

Gets or sets the coefficients of the polynomial.

compute_covariance_matrix(function, best_fit_parameters)

Compute the covariance matrix of this fit :param function: the loglike for the fit :param best_fit_parameters: the best fit parameters :return:

property covariance_matrix

property degree

the polynomial degree :return:

property error

the error on the polynomial coefficients :return:

classmethod from_previous_fit(coefficients, covariance)

integral(xmin, xmax)

Evaluate the integral of the polynomial between xmin and xmax

integral_error(xmin, xmax)

computes the integral error of an interval :param xmin: start of the interval :param xmax: stop of the interval :return: interval error

threeML.utils.time_series.polynomial.polyfit(x, y, grade, exposure)

function to fit a polynomial to event data. not a member to allow parallel computation

threeML.utils.time_series.polynomial.unbinned_polyfit(events, grade, t_start, t_stop, exposure, initial_amplitude=1)

function to fit a polynomial to event data. not a member to allow parallel computation

threeML.utils.time_series.time_series module

exception threeML.utils.time_series.time_series.OverLappingIntervals

Bases: RuntimeError

exception threeML.utils.time_series.time_series.ReducingNumberOfSteps

Bases: Warning

exception threeML.utils.time_series.time_series.ReducingNumberOfThreads

Bases: Warning

class threeML.utils.time_series.time_series.TimeSeries(start_time, stop_time, n_channels, native_quality=None, first_channel=1, ra=None, dec=None, mission=None, instrument=None, verbose=True, edges=None)

Bases: object

property bins

count_per_channel_over_interval(start, stop)

Parameters

• start –

• stop –

Returns

counts_over_interval(start, stop)

return the number of counts in the selected interval :param start: start of interval :param stop: stop of interval :return:

exposure_over_interval(tmin, tmax)

calculate the exposure over a given interval

get_information_dict(use_poly=False, extract=False)

Return a PHAContainer that can be read by different builders

Parameters

use_poly – (bool) choose to build from the polynomial fits

get_poly_info()

Return a pandas panel frame with the polynomial coeffcients and errors Returns:

a DataFrame

get_total_poly_count(start, stop, mask=None)

Get the total poly counts

Parameters

• start –

• stop –

Returns

get_total_poly_error(start, stop, mask=None)

Get the total poly error

Parameters

• start –

• stop –

Returns

property n_channels

property poly_fit_exists

property poly_intervals

property poly_order

Get or set the polynomial order

property polynomials

Returns polynomial is they exist

restore_fit(filename)

save_background(filename, overwrite=False)

save the background to an HD5F

Parameters

filename –

Returns

set_active_time_intervals(*args)

set_polynomial_fit_interval(*time_intervals, **options)

Set the time interval to fit the background. Multiple intervals can be input as separate arguments Specified as ‘tmin-tmax’. Intervals are in seconds. Example:

set_polynomial_fit_interval(“-10.0-0.0”,”10.-15.”)

Parameters

• time_intervals – intervals to fit on

• options –

property time_intervals

the time intervals of the events

Returns

view_lightcurve(start=- 10, stop=20.0, dt=1.0, use_binner=False)

threeML.utils.time_series.time_series.ceildiv(a, b)

Module contents