mne_denoise.ssa.SingularSpectrumAnalysis#

class mne_denoise.ssa.SingularSpectrumAnalysis(sfreq: float | None = None, window_length: int | None = None, drop_freq_max: float = 3.0, drop_band: tuple[float, float] | None = None, n_check: int | None = None, max_window: int = 100, verbose: bool | str | int | None = None, *, window_seconds: float | None = None)[source]#

Frequency-guided per-channel Basic SSA transformer.

Parameters:

  • sfreq (float | None, default=None) – Sampling frequency in Hz. NumPy input requires an explicit value. MNE input supplies it from metadata and must agree with an explicit value.

  • window_length (int | None, default=None) – Embedding dimension in samples. It is mutually exclusive with window_seconds. None selects an automatic value.

  • drop_freq_max (float, default=3.0) – Reject components whose dominant frequency is at or below this value in Hz.

  • drop_band (tuple of float | None, default=None) – Inclusive (low, high) dominant-frequency rejection band in Hz. If supplied, it replaces drop_freq_max as the component-selection interval.

  • n_check (int | None, default=None) – Restrict selection to this many leading numerical-rank components. None examines every numerical-rank component.

  • max_window (int, default=100) – Maximum embedding dimension used by automatic window selection.

  • verbose (bool | str | int | None, default=None) – MNE-style logging level.

  • window_seconds (float | None, default=None) – Embedding duration in seconds, mutually exclusive with window_length.

sfreq_#

Validated sampling frequency used during fitting.

Type:

float

n_channels_in_#

Number of data channels seen during fitting.

Type:

int

ch_names_in_#

Fitted MNE channel names and order, or None for NumPy input.

Type:

tuple of str | None

diagnostics_#

Diagnostics from the most recent transformation. Epoched input stores one dictionary per epoch.

Type:

dict | list of dict

dropped_counts_#

Number of rejected components per channel, or per epoch and channel.

Type:

ndarray

dropped_frequencies_#

Dominant frequencies of rejected components for every channel.

Type:

list

See also

compute_basic_ssa

Functional interface for channel-first arrays.

ssa_clean_channel

Canonical single-channel implementation.

ssa_decompose

Complete additive Basic SSA decomposition.

mne_denoise.ssa.LocalSingularSpectrumAnalysis

Local clustered SSA.

Notes

The estimator is transductive. fit validates the operating point and records the channel layout; every transform decomposes the records supplied to that call. Changing record or epoch boundaries can therefore change the trajectory matrix, Fourier bins, and selected components. The additive decomposition follows Basic SSA [1]; dominant-frequency rejection is an application-specific grouping rule.

Examples

>>> import numpy as np
>>> from mne_denoise.ssa import SingularSpectrumAnalysis
>>> sfreq = 100.0
>>> time = np.arange(500) / sfreq
>>> data = np.vstack(
...     [np.sin(2 * np.pi * 1.0 * time), np.sin(2 * np.pi * 10.0 * time)]
... )
>>> model = SingularSpectrumAnalysis(sfreq=sfreq, drop_freq_max=3.0)
>>> cleaned = model.fit_transform(data)
>>> cleaned.shape
(2, 500)

References

[1]

Golyandina, N., & Zhigljavsky, A. (2013). Singular Spectrum Analysis for Time Series. Springer. https://doi.org/10.1007/978-3-642-34913-3

__init__(sfreq: float | None = None, window_length: int | None = None, drop_freq_max: float = 3.0, drop_band: tuple[float, float] | None = None, n_check: int | None = None, max_window: int = 100, verbose: bool | str | int | None = None, *, window_seconds: float | None = None) → None[source]#

Methods

__init__([sfreq, window_length, ...])

fit(X[, y])

Validate the operating point and record the fitted channel layout.

fit_transform(X[, y])

Fit to data, then transform it.

get_metadata_routing()

Get metadata routing of this object.

get_params([deep])

Get parameters for this estimator.

set_output(*[, transform])

Set output container.

set_params(**params)

Set the parameters of this estimator.

transform(X[, y])

Apply the transductive SSA decomposition to the supplied records.