xrscipy.signal.extra.frequency_filter

xrscipy.signal.extra.frequency_filter(darray: DataArray, f_crit: _SupportsArray[dtype[Any]] | _NestedSequence[_SupportsArray[dtype[Any]]] | bool | int | float | complex | str | bytes | _NestedSequence[bool | int | float | complex | str | bytes], order: int = None, irtype: str = 'iir', filtfilt: bool = True, apply_kwargs: dict = None, in_nyq: bool = False, dim: str = None, **kwargs) → DataArray

Applies given frequency filter to a darray.

This is a 1-d filter. If the darray is one dimensional, then the dimension along which the filter is applied is chosen automatically if not specified by dim. If darray is multi dimensional then axis along which the filter is applied has to be specified by dim string.

The type of the filter is chosen by irtype and then filtfilt states is the filter is applied both ways, forward and backward. Additional parameters passed to filter function specified by apply_kwargs.

If ‘iir’ is chosen as irtype, then if filtfilt is True then the filter scipy.signal.filtfilt is used, if False scipy.signal.lfilter applies.

If ‘fir’ is chosen as irtype, then if filtfilt is True then the filter scipy.signal.sosfiltfilt is used, if False scipy.signal.sosfilt applies.

Parameters:

  • darray (DataArray) – An xarray type data to be filtered.

  • f_crit (array_like) – A scalar or length-2 sequence giving the critical frequencies.

  • order (int, optional) – The order of the filter. If Default then it takes order defaults from _ORDER_DEFAULTS, which is irtype specific. Default is None.

  • irtype (string, optional) – A string specifying the impulse response of the filter, has to be either “fir” then finite impulse response (FIR) is used, or “iir” then infinite impulse response (IIR) filter is applied. ValueError is raised otherwise. Default is “iir”.

  • filtfilt (bool, optional) – When True the filter is applied both forwards and backwards, otherwise only one way, from left to right, is applied. Default is True.

  • apply_kwargs (dict, optional) – Specifies kwargs, which are passed to the filter function given by irtype and filtfilt. Default is None.

  • in_nyq (bool, optional) – If True, then the critical frequencies given by f_crit are normalized by Nyquist frequency. Default is False.

  • dim (string, optional) – A string specifing the dimension along which the filter is applied. If darray is 1-d then the dimension is found if not specified by dim. For multi dimensional darray has to be specified, otherwise ValueError is raised. Default is None.

  • kwargs – Arbitrary keyword arguments passed when the filter is being designed, either to scipy.signal.iirfilter if “iir” method for irtype is chosen, or scipy.signal.firwin.