pyrasa.utils.fit_funcs.AbstractFitFun

class pyrasa.utils.fit_funcs.AbstractFitFun(freq: ndarray, aperiodic_spectrum: ndarray, scale_factor: int | float)

Abstract base class for fitting functions used to model aperiodic spectra.

This class provides a framework for defining and fitting models to aperiodic spectra. It handles common functionality required for fitting a model, such as scaling and goodness-of-fit computation. Subclasses should implement the func method to define the specific fitting function used for curve fitting.

freq

The frequency values associated with the aperiodic spectrum data.

Type:

np.ndarray

aperiodic_spectrum

The aperiodic spectrum data to which the model will be fit.

Type:

np.ndarray

scale_factor

A scaling factor used to adjust the fit results.

Type:

int | float

label

A label to identify the type of fit or model used. Default is ‘custom’.

Type:

ClassVar[str]

log10_aperiodic

If True, the aperiodic spectrum values will be transformed using log10. Default is False.

Type:

ClassVar[bool]

log10_freq

If True, the frequency values will be transformed using log10. Default is False.

Type:

ClassVar[bool]

__attrs_post_init__()

Post-initialization method to apply log10 transformations if specified.

func(x: np.ndarray, \*args: float) → np.ndarray

Abstract method to define the model function. Must be implemented by subclasses and should be applicable to scipy.optimize.curve_fit.

curve_kwargs() → dict[str, Any]

Returns keyword arguments for the curve fitting process.

add_infos_to_df(df_params: pd.DataFrame) → pd.DataFrame

Method to add additional information to the parameters DataFrame. Can be overridden by subclasses.

handle_scaling(df_params: pd.DataFrame, scale_factor: float) → pd.DataFrame

Adjusts the parameters DataFrame based on the scaling factor. Can be overridden by subclasses.

fit_func() → tuple[pd.DataFrame, pd.DataFrame]

Fits the model to the data and returns DataFrames containing the model parameters and goodness-of-fit metrics.

Notes

• Subclasses must implement the func method to define the model’s functional form.

• The curve_kwargs method can be overridden to customize curve fitting options.

• The add_infos_to_df and handle_scaling methods are intended to be overridden if additional functionality or specific scaling behavior is required.

References

For details on goodness-of-fit metrics and their calculations, see the documentation for _get_gof.

Attributes:

aperiodic_spectrum

curve_kwargs

freq

scale_factor

Methods

add_infos_to_df
fit_func
func
handle_scaling

__hash__ = None