ada.long.cosinor#

Classes#

`Antilogistic`	A class for fitting anti-logistic transform of a sine wave, with asymetric and unknown base period using nonlinear methods.
`Arctangent`	A class for fitting arctangent transform of a sine wave, with asymetric and unknown base period using nonlinear methods.
`CosinorResults`	A container for storing results of cosinor model fitting. Do not construct manually! Fit params gives access to the following:
`Hill`	A class for fitting Hill transform of a sine wave, with asymetric and unknown base period using nonlinear methods.
`Linear`	A class for fitting linear combination of harmonic sines of known base period to the data using Ordinary Least Squares.
`Nonlinear`	A class for fitting linear combination of harmonic sines of unknown base period to the data using nonlinear methods.
`Saw`	A class for fitting saw wave, possibly assymetric, of unknown base period to the data using nonlinear methods.

Module Contents#

class Antilogistic(test_periods=(16.0, 32.0), preprocessing=False, use_bootstrap=False)[source]#

A class for fitting anti-logistic transform of a sine wave, with asymetric and unknown base period using nonlinear methods.

test_periods#

Return type:: list[float]

Lower and upper bounds for fitting periods, in hours. Defaults to (16, 32).

Type:

Tuple[float, float]

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)

preprocessing#

Return type:: bool

If true, data will be detrended and low-pass filtered before applying cosinor. Defaults to False.

Type:

bool

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)

use_bootstrap#

Return type:: bool

If true, significance of fitted amplitude will be assessed using bootstrap. Might take some time for high sampling frequencies. Defaults to False.

Type:

bool

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)

fit(data, ch_name=None)#

Fits the specified by the object curve to provided data.

Parameters:

data (_ActiData) – Actigraphic data to be fitted with cosinor method.
ch_name (str | None, optional) – Channel to which curve will be fitted. If None, it will be to_score in case of Raw or Epoched data and score in case of ScoredShort data. Defaults to None.

Returns:

List of objects containing fitting results. It containes only 1 element.

Return type:

list[CosinorResults]

property preprocessing: bool#

Whether preprocessing is applied.

Return type:: bool

property test_periods: list[float]#

List of periods, that will be fitted to the data.

Return type:: list[float]

property use_bootstrap: bool#

Whether bootstrap method is used to compute p-values of amplitudes.

Return type:: bool

class Arctangent(test_periods=(16, 32), preprocessing=False, use_bootstrap=False)[source]#

A class for fitting arctangent transform of a sine wave, with asymetric and unknown base period using nonlinear methods.

test_periods#

Return type:: list[float]

Lower and upper bounds for fitting periods, in hours. Defaults to (16, 32).

Type:

Tuple[float, float]

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)

preprocessing#

Return type:: bool

If true, data will be detrended and low-pass filtered before applying cosinor. Defaults to False.

Type:

bool

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)

use_bootstrap#

Return type:: bool

If true, significance of fitted amplitude will be assessed using bootstrap. Might take some time for high sampling frequencies. Defaults to False.

Type:

bool

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)

fit(data, ch_name=None)#

Fits the specified by the object curve to provided data.

Parameters:

data (_ActiData) – Actigraphic data to be fitted with cosinor method.
ch_name (str | None, optional) – Channel to which curve will be fitted. If None, it will be to_score in case of Raw or Epoched data and score in case of ScoredShort data. Defaults to None.

Returns:

List of objects containing fitting results. It containes only 1 element.

Return type:

list[CosinorResults]

property preprocessing: bool#

Whether preprocessing is applied.

Return type:: bool

property test_periods: list[float]#

List of periods, that will be fitted to the data.

Return type:: list[float]

property use_bootstrap: bool#

Whether bootstrap method is used to compute p-values of amplitudes.

Return type:: bool

class CosinorResults(sine_params, period, amplitude, pvals, method, ch, preprocessing, id, generalized_params=None)[source]#

A container for storing results of cosinor model fitting. Do not construct manually! Fit params gives access to the following:

base freq, n harmonic indexed from 1: dicts containing keys: period, amplitude, phase
amplitude: summary amplitude of fitted line
constant: MESOR
f-test: pvalue from F-test
bootstrap: pvalue from bootstrap or None
method: used model
preprocessing: whether it was used
channel: to which the model was fit

Parameters:

sine_params (numpy.ndarray)
period (float)
amplitude (float)
pvals (Tuple[float, float | None])
method (str)
ch (str | None)
preprocessing (bool)
id (str)
generalized_params (dict | None)

export(out_path)[source]#

Save data (parameters of the model) to compressed generic file.

Parameters:: out_path (str) – Path to the out file.

static load_file(path)[source]#

Load model fitting results from .ada.long file (created via export method).

Parameters:: path (str) – Path to file.
Returns:: Object containing fitted model.
Return type:: CosinorResults

plot(out_path, data)[source]#

Plot the provided data together with fitted model.

Parameters:

out_path (str | None) – Path to save the plot. If None, the plot will be opened in interactive window. Defaults to None.
data (_ActiData) – Data to which the model was fit.

save_csv(out_path)[source]#

Save parameters of fitted model to human-readable csv.

Parameters:: out_path (str) – Path to the out file.

property fit_params: dict#

Parameters of the fitted model.

Return type:: dict

property id: str#

ID of a recording to which long estimate was fit.

Return type:: str

class Hill(test_periods=(16, 32), preprocessing=False, use_bootstrap=False)[source]#

A class for fitting Hill transform of a sine wave, with asymetric and unknown base period using nonlinear methods.

test_periods#

Return type:: list[float]

Lower and upper bounds for fitting periods, in hours. Defaults to (16, 32).

Type:

Tuple[float, float]

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)

preprocessing#

Return type:: bool

If true, data will be detrended and low-pass filtered before applying cosinor. Defaults to False.

Type:

bool

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)

use_bootstrap#

Return type:: bool

If true, significance of fitted amplitude will be assessed using bootstrap. Might take some time for high sampling frequencies. Defaults to False.

Type:

bool

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)

fit(data, ch_name=None)#

Fits the specified by the object curve to provided data.

Parameters:

data (_ActiData) – Actigraphic data to be fitted with cosinor method.
ch_name (str | None, optional) – Channel to which curve will be fitted. If None, it will be to_score in case of Raw or Epoched data and score in case of ScoredShort data. Defaults to None.

Returns:

List of objects containing fitting results. It containes only 1 element.

Return type:

list[CosinorResults]

property preprocessing: bool#

Whether preprocessing is applied.

Return type:: bool

property test_periods: list[float]#

List of periods, that will be fitted to the data.

Return type:: list[float]

property use_bootstrap: bool#

Whether bootstrap method is used to compute p-values of amplitudes.

Return type:: bool

class Linear(test_periods=24, preprocessing=False, use_bootstrap=False, n_components=1)[source]#

A class for fitting linear combination of harmonic sines of known base period to the data using Ordinary Least Squares.

test_periods#

Return type:: list[float]

Periods in hours to be fitted to the data. Defaults to 24.

Type:

list[float] | float

Parameters:

test_periods (list[float] | float)
preprocessing (bool)
use_bootstrap (bool)
n_components (int)

preprocessing#

Return type:: bool

If true, data will be detrended and low-pass filtered before applying cosinor. Defaults to False.

Type:

bool

Parameters:

test_periods (list[float] | float)
preprocessing (bool)
use_bootstrap (bool)
n_components (int)

use_bootstrap#

Return type:: bool

If true, significance of fitted amplitude will be assessed using bootstrap. Might take some time for high sampling frequencies. Defaults to False.

Type:

bool

Parameters:

test_periods (list[float] | float)
preprocessing (bool)
use_bootstrap (bool)
n_components (int)

n_components#

Return type:: int

Number of harmonics, that will be present in the fitted curve. Defaults to 1 (no harmonics, only base frequency).

Type:

int

Parameters:

test_periods (list[float] | float)
preprocessing (bool)
use_bootstrap (bool)
n_components (int)

fit(data, ch_name=None)[source]#

Fits the specified by the object curves to provided data.

Parameters:

data (_ActiData) – Actigraphic data to be fitted with cosinor method.
ch_name (str | None, optional) – Channel to which curve will be fitted. If None, it will be to_score in case of Raw or Epoched data and score in case of ScoredShort data. Defaults to None.

Returns:

List of objects containing fitting results. Each element corresponds to one period.

Return type:

list[CosinorResults]

property n_components: int#

Number of harmonic components fitted to the data.

Return type:: int

property preprocessing: bool#

Whether preprocessing is applied.

Return type:: bool

property test_periods: list[float]#

List of periods, that will be fitted to the data.

Return type:: list[float]

property use_bootstrap: bool#

Whether bootstrap method is used to compute p-values of amplitudes.

Return type:: bool

class Nonlinear(test_periods=(16, 32), preprocessing=False, use_bootstrap=False, n_components=1)[source]#

A class for fitting linear combination of harmonic sines of unknown base period to the data using nonlinear methods.

test_periods#

Return type:: list[float]

Lower and upper bounds for fitting periods, in hours. Defaults to (16, 32).

Type:

Tuple[float, float]

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)
n_components (int)

preprocessing#

Return type:: bool

If true, data will be detrended and low-pass filtered before applying cosinor. Defaults to False.

Type:

bool

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)
n_components (int)

use_bootstrap#

Return type:: bool

If true, significance of fitted amplitude will be assessed using bootstrap. Might take some time for high sampling frequencies. Defaults to False.

Type:

bool

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)
n_components (int)

n_components#

Return type:: int

Number of harmonics, that will be present in the fitted curve. Defaults to 1 (no harmonics, only base frequency).

Type:

int

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)
n_components (int)

fit(data, ch_name=None)[source]#

Fits the specified by the object curves to provided data.

Parameters:

data (_ActiData) – Actigraphic data to be fitted with cosinor method.
ch_name (str | None, optional) – Channel to which curve will be fitted. If None, it will be to_score in case of Raw or Epoched data and score in case of ScoredShort data. Defaults to None.

Returns:

List of objects containing fitting results. It containes only 1 element.

Return type:

list[CosinorResults]

property n_components: int#

Number of harmonic components fitted to the data.

Return type:: int

property preprocessing: bool#

Whether preprocessing is applied.

Return type:: bool

property test_periods: list[float]#

List of periods, that will be fitted to the data.

Return type:: list[float]

property use_bootstrap: bool#

Whether bootstrap method is used to compute p-values of amplitudes.

Return type:: bool

class Saw(test_periods=(16, 32), preprocessing=False, use_bootstrap=False)[source]#

A class for fitting saw wave, possibly assymetric, of unknown base period to the data using nonlinear methods.

test_periods#

Return type:: list[float]

Lower and upper bounds for fitting periods, in hours. Defaults to (16, 32).

Type:

Tuple[float, float]

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)

preprocessing#

Return type:: bool

If true, data will be detrended and low-pass filtered before applying cosinor. Defaults to False.

Type:

bool

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)

use_bootstrap#

Return type:: bool

If true, significance of fitted amplitude will be assessed using bootstrap. Might take some time for high sampling frequencies. Defaults to False.

Type:

bool

Parameters:

test_periods (Tuple[float, float])
preprocessing (bool)
use_bootstrap (bool)

fit(data, ch_name=None)#

Fits the specified by the object curve to provided data.

Parameters:

data (_ActiData) – Actigraphic data to be fitted with cosinor method.
ch_name (str | None, optional) – Channel to which curve will be fitted. If None, it will be to_score in case of Raw or Epoched data and score in case of ScoredShort data. Defaults to None.

Returns:

List of objects containing fitting results. It containes only 1 element.

Return type:

list[CosinorResults]

property preprocessing: bool#

Whether preprocessing is applied.

Return type:: bool

property test_periods: list[float]#

List of periods, that will be fitted to the data.

Return type:: list[float]

property use_bootstrap: bool#

Whether bootstrap method is used to compute p-values of amplitudes.

Return type:: bool