pymultifracs.cumulants.Cumulants#

class pymultifracs.cumulants.Cumulants(mrq: dataclasses.InitVar[MultiResolutionQuantity], n_cumul: int, scaling_ranges: ~typing.List[~typing.Tuple[int]], bootstrapped_mfa: dataclasses.InitVar[MFractalVar] = None, weighted: str = None, robust_kwargs: ~typing.Dict[str, object] = <factory>, robust: dataclasses.InitVar[bool] = False)#

Computes and analyzes cumulants

Parameters:

mrqMultiResolutionQuantity: Multi resolution quantity to analyze.
n_cumulint: Number of cumulants to compute.
j1int: Lower-bound of the scale support for the linear regressions.
j2int: Upper-bound of the scale support for the linear regressions.
weighted: str | None: Whether to used weighted linear regressions.

Attributes:

formalismstr: Formalism used. Can be any of ‘wavelet coefs’, ‘wavelet leaders’, or ‘wavelet p-leaders’.
njdict(ndarray): Number of coefficients at scale j. Arrays are of the shape (n_rep,)
valuesndarray, shape (n_cumulants, n_scales, n_rep): \(C_m(j)\).
n_cumulint: Number of computed cumulants.
j1int: Lower-bound of the scale support for the linear regressions.
j2int: Upper-bound of the scale support for the linear regressions.
weightedstr | None: Whether weighted regression was performed.
mndarray, shape (n_cumulants,): List of the m values (cumulants), in order presented in the value arrays.
jndarray, shape (n_scales,): List of the j values (scales), in order presented in the value arrays.
log_cumulantsndarray, shape (n_cumulants, n_rep): \((c_m)_m\), slopes of the curves \(j \times C_m(j)\).
var_log_cumulantsndarray, shape (n_cumulants, n_rep): Estimates of the log-cumulants

Warning

var_log_cumulants was not debugged
n_repint: Number of realisations

classmethod from_dict(d)#

Method to instanciate a dataclass by passing a dictionary with extra keywords

Parameters:

ddict: Dictionary containing at least all the parameters required by __init__, but can also contain other parameters, which will be ignored

Returns:

MultiResolutionQuantityBase: Properly initialized multi resolution quantity

Notes

Note

Normally, dataclasses can only be instantiated by only specifiying parameters expected by the automatically generated __init__ method. Using this method instead allows us to discard extraneous parameters, similarly to introducing a **kwargs parameter.

get_nj()#: Returns nj as a list

get_nj_interv(j1, j2)#: Returns nj as an array, for j in [j1,j2]