mne.filter.filter_data#
- mne.filter.filter_data(data, sfreq, l_freq, h_freq, picks=None, filter_length='auto', l_trans_bandwidth='auto', h_trans_bandwidth='auto', n_jobs=None, method='fir', iir_params=None, copy=True, phase='zero', fir_window='hamming', fir_design='firwin', pad='reflect_limited', *, verbose=None)[source]#
Filter a subset of channels.
- Parameters:
- data
ndarray
, shape (…, n_times) The data to filter.
- sfreq
float
The sample frequency in Hz.
- l_freq
float
|None
For FIR filters, the lower pass-band edge; for IIR filters, the lower cutoff frequency. If None the data are only low-passed.
- h_freq
float
|None
For FIR filters, the upper pass-band edge; for IIR filters, the upper cutoff frequency. If None the data are only high-passed.
- picks
list
|slice
|None
Channels to include. Slices and lists of integers will be interpreted as channel indices. None (default) will pick all channels. Note that channels in
info['bads']
will be included if their indices are explicitly provided. Currently this is only supported for 2D (n_channels, n_times) and 3D (n_epochs, n_channels, n_times) arrays.- filter_length
str
|int
Length of the FIR filter to use (if applicable):
‘auto’ (default): The filter length is chosen based on the size of the transition regions (6.6 times the reciprocal of the shortest transition band for fir_window=’hamming’ and fir_design=”firwin2”, and half that for “firwin”).
str: A human-readable time in units of “s” or “ms” (e.g., “10s” or “5500ms”) will be converted to that number of samples if
phase="zero"
, or the shortest power-of-two length at least that duration forphase="zero-double"
.int: Specified length in samples. For fir_design=”firwin”, this should not be used.
- l_trans_bandwidth
float
|str
Width of the transition band at the low cut-off frequency in Hz (high pass or cutoff 1 in bandpass). Can be “auto” (default) to use a multiple of
l_freq
:min(max(l_freq * 0.25, 2), l_freq)
Only used for
method='fir'
.- h_trans_bandwidth
float
|str
Width of the transition band at the high cut-off frequency in Hz (low pass or cutoff 2 in bandpass). Can be “auto” (default in 0.14) to use a multiple of
h_freq
:min(max(h_freq * 0.25, 2.), info['sfreq'] / 2. - h_freq)
Only used for
method='fir'
.- n_jobs
int
|str
Number of jobs to run in parallel. Can be
'cuda'
ifcupy
is installed properly andmethod='fir'
.- method
str
'fir'
will use overlap-add FIR filtering,'iir'
will use IIR forward-backward filtering (viafiltfilt()
).- iir_params
dict
|None
Dictionary of parameters to use for IIR filtering. If
iir_params=None
andmethod="iir"
, 4th order Butterworth will be used. For more information, seemne.filter.construct_iir_filter()
.- copybool
If True, a copy of x, filtered, is returned. Otherwise, it operates on x in place.
- phase
str
Phase of the filter. When
method='fir'
, symmetric linear-phase FIR filters are constructed with the following behaviors whenmethod="fir"
:"zero"
(default)The delay of this filter is compensated for, making it non-causal.
"minimum"
A minimum-phase filter will be constructed by decomposing the zero-phase filter into a minimum-phase and all-pass systems, and then retaining only the minimum-phase system (of the same length as the original zero-phase filter) via
scipy.signal.minimum_phase()
."zero-double"
This is a legacy option for compatibility with MNE <= 0.13. The filter is applied twice, once forward, and once backward (also making it non-causal).
"minimum-half"
This is a legacy option for compatibility with MNE <= 1.6. A minimum-phase filter will be reconstructed from the zero-phase filter with half the length of the original filter.
When
method='iir'
,phase='zero'
(default) or equivalently'zero-double'
constructs and applies IIR filter twice, once forward, and once backward (making it non-causal) usingfiltfilt()
;phase='forward'
will apply the filter once in the forward (causal) direction usinglfilter()
.New in v0.13.
Changed in version 1.7: The behavior for
phase="minimum"
was fixed to use a filter of the requested length and improved suppression.- fir_window
str
The window to use in FIR design, can be “hamming” (default), “hann” (default in 0.13), or “blackman”.
New in v0.15.
- fir_design
str
Can be “firwin” (default) to use
scipy.signal.firwin()
, or “firwin2” to usescipy.signal.firwin2()
. “firwin” uses a time-domain design technique that generally gives improved attenuation using fewer samples than “firwin2”.New in v0.15.
- pad
str
The type of padding to use. Supports all
numpy.pad()
mode
options. Can also be"reflect_limited"
, which pads with a reflected version of each vector mirrored on the first and last values of the vector, followed by zeros. Only used formethod='fir'
. The default is'reflect_limited'
.New in v0.15.
- verbosebool |
str
|int
|None
Control verbosity of the logging output. If
None
, use the default verbosity level. See the logging documentation andmne.verbose()
for details. Should only be passed as a keyword argument.
- data
- Returns:
- data
ndarray
, shape (…, n_times) The filtered data.
- data
Notes
Applies a zero-phase low-pass, high-pass, band-pass, or band-stop filter to the channels selected by
picks
.l_freq
andh_freq
are the frequencies below which and above which, respectively, to filter out of the data. Thus the uses are:l_freq < h_freq
: band-pass filterl_freq > h_freq
: band-stop filterl_freq is not None and h_freq is None
: high-pass filterl_freq is None and h_freq is not None
: low-pass filter
Note
If n_jobs > 1, more memory is required as
len(picks) * n_times
additional time points need to be temporarily stored in memory.For more information, see the tutorials Background information on filtering and Filtering and resampling data and
mne.filter.create_filter()
.