API reference

ASR

mne_denoise.asr.ASR	Artifact Subspace Reconstruction (ASR) scikit-learn transformer.
mne_denoise.asr.AdaptiveASR	Adaptive Artifact Subspace Reconstruction (AASR) estimator.
mne_denoise.asr.JugglerASR	Juggler Artifact Subspace Reconstruction (JugglerASR).
mne_denoise.asr.GuidedASR	DSS-biased soft Artifact Subspace Reconstruction (experimental).
mne_denoise.asr.calibrate_asr	Calibrate a standard ASR model from continuous data.
mne_denoise.asr.compute_clean_window_mask	Compute a statistical retained-sample mask for continuous data.
mne_denoise.asr.fit_rms_distribution	Fit robust clean EEG RMS statistics.
mne_denoise.asr.process_asr	Apply a calibrated ASR model to continuous data.
mne_denoise.asr.process_guided_asr	Apply a calibrated ASR state with guided soft reconstruction.
mne_denoise.asr.select_juggler_reference_samples	Select calibration samples using Juggler's ASR rules.

Warning

GuidedASR and process_guided_asr are unpublished, unvalidated experimental research prototypes. Their current evidence is limited to unit tests and synthetic benchmarks; independently validate signal preservation and artifact attenuation before scientific use.

DSS

mne_denoise.dss.compute_dss	Compute DSS spatial filters from baseline and biased covariances.
mne_denoise.dss.DSS	Denoising Source Separation (DSS) Transformer.
mne_denoise.dss.TimeShiftDSS	Trial-average DSS in a lag-augmented sensor space.
mne_denoise.dss.iterative_dss	Extract multiple DSS components using iterative (nonlinear) algorithm.
mne_denoise.dss.IterativeDSS	Iterative (Nonlinear) Denoising Source Separation Transformer.

ZapLine

mne_denoise.zapline.ZapLine

ZapLine Transformer for line noise removal.

Spectrum interpolation

mne_denoise.spectrum_interpolation.SpectrumInterpolation	Remove power-line noise by amplitude spectrum interpolation.
mne_denoise.spectrum_interpolation.interpolate_spectrum	Remove line noise from 2D data by amplitude spectrum interpolation.

iCanClean

mne_denoise.icanclean.ICanClean	ICanClean Transformer for reference-based artifact removal.
mne_denoise.icanclean.compute_icanclean	Compute one iCanClean pass on continuous NumPy arrays.

BSS-CCA

mne_denoise.bss_cca.BSSCCA	Reference-free BSS-CCA artifact-attenuation estimator.
mne_denoise.bss_cca.compute_bss_cca	Learn and apply reference-free BSS-CCA to a channel-first array.

SNS

mne_denoise.sns.SNS	Sensor Noise Suppression estimator.
mne_denoise.sns.compute_sns	Learn and apply Sensor Noise Suppression to a channel-first array.
mne_denoise.sns.compute_sns_weights	Compute the SNS spatial operator from a channel covariance matrix.

SSA

mne_denoise.ssa.SingularSpectrumAnalysis	Frequency-guided per-channel Basic SSA transformer.
mne_denoise.ssa.LocalSingularSpectrumAnalysis	Local-SSA high-amplitude artifact transformer.
mne_denoise.ssa.ssa_decompose	Decompose a one-dimensional series into additive Basic SSA components.
mne_denoise.ssa.ssa_w_correlation	Compute weighted correlations between SSA reconstructions.
mne_denoise.ssa.compute_basic_ssa	Apply frequency-guided Basic SSA independently to every channel.
mne_denoise.ssa.ssa_clean_channel	Clean one channel by grouping Basic SSA components by frequency.
mne_denoise.ssa.compute_local_ssa	Apply local SSA independently to every input channel.
mne_denoise.ssa.local_ssa_clean_channel	Remove a locally reconstructed high-energy artifact from one channel.

Denoisers

mne_denoise.dss.denoisers.LinearDenoiser	Base class for linear bias functions.
mne_denoise.dss.denoisers.AverageBias	Bias function for finding repeatable components via averaging.
mne_denoise.dss.denoisers.CycleAverageBias	Fixed-window event-locked averaging bias.
mne_denoise.dss.denoisers.BandpassBias	Bandpass filter bias for narrow-band rhythm extraction.
mne_denoise.dss.denoisers.LineNoiseBias	A bias LinearDenoiser for line noise isolation (Notch/IIR or FFT/Harmonic).
mne_denoise.dss.denoisers.PeakFilterBias	Peak filter bias for single-frequency extraction.
mne_denoise.dss.denoisers.CombFilterBias	Comb filter bias for harmonic frequency extraction.
mne_denoise.dss.denoisers.LagAverageBias	Lag-averaging bias for emphasizing temporally smooth signals.
mne_denoise.dss.denoisers.SmoothingBias	Unified temporal smoothing bias (Moving Average).
mne_denoise.dss.denoisers.SpectrogramBias	Linear spectrogram bias (Section 4.1.3).
mne_denoise.dss.denoisers.NonlinearDenoiser	Base class for nonlinear/adaptive denoiser functions.
mne_denoise.dss.denoisers.TanhMaskDenoiser	Tanh mask denoiser (Standard FastICA nonlinearity).
mne_denoise.dss.denoisers.RobustTanhDenoiser	Robust tanh denoiser (FastICA / RobustICA formulation).
mne_denoise.dss.denoisers.KurtosisDenoiser	Kurtosis maximization denoiser.
mne_denoise.dss.denoisers.SkewDenoiser	Skewness nonlinearity (FastICA 'skew').
mne_denoise.dss.denoisers.GaussDenoiser	Gaussian nonlinearity (FastICA 'gauss').
mne_denoise.dss.denoisers.WienerMaskDenoiser	Adaptive Wiener mask denoiser.
mne_denoise.dss.denoisers.SpectrogramDenoiser	Adaptive/Nonlinear spectrogram denoiser (Section 4.1.3).
mne_denoise.dss.denoisers.DCTDenoiser	DCT domain denoiser (MATLAB denoise_dct.m).
mne_denoise.dss.denoisers.QuasiPeriodicDenoiser	Quasi-periodic denoiser via cycle averaging.

Variants

mne_denoise.dss.variants.smooth_dss	Create an ordinary DSS configured for temporally smooth sources.
mne_denoise.dss.variants.narrowband_dss	Create a DSS configured for a specific frequency band.
mne_denoise.dss.variants.narrowband_scan	Scan frequencies to find optimal narrowband DSS components.
mne_denoise.dss.variants.ssvep_dss	Create a DSS configured for SSVEP extraction.

Quality Assurance

mne_denoise.qa.peak_attenuation_db	Attenuation (dB) of the dominant peak around a target frequency.
mne_denoise.qa.suppression_ratio	Suppression ratio (dB) of mean band power around a target frequency.
mne_denoise.qa.noise_surround_ratio	Residual peak-to-surround power ratio around a target frequency.
mne_denoise.qa.below_noise_distortion_db	Broadband spectral distortion (dB) outside excluded noise bands.
mne_denoise.qa.spectral_distortion	Spectral distortion (dB RMS) at non-harmonic frequencies.
mne_denoise.qa.overclean_proportion	Fraction of channels where the spectral floor is over-suppressed.
mne_denoise.qa.underclean_proportion	Fraction of channels where the line-noise peak remains prominent.
mne_denoise.qa.geometric_mean_psd_ratio	Geometric mean of psd_after / psd_before across broadband.
mne_denoise.qa.variance_removed	Percentage of total variance removed after denoising.
mne_denoise.qa.compute_all_qa_metrics	Compute all QA metrics for a line-noise removal benchmark.
mne_denoise.qa.rms_change	Root mean square (RMS) of the difference between before and after signals.
mne_denoise.qa.max_abs_change	Maximum absolute change between before and after signals.
mne_denoise.qa.channel_variance_ratio	Per-channel variance ratio: var(after) / var(before).

Visualization

mne_denoise.viz.plot_asr_repair_timeline	Plot the per-window count of reconstructed components over time.
mne_denoise.viz.plot_asr_calibration_fraction	Bar chart of the clean calibration fraction for one or more estimators.
mne_denoise.viz.plot_asr_component_reconstruction	Heatmap of per-window component variance relative to rejection thresholds.
mne_denoise.viz.plot_guided_asr_weights	Plot experimental GuidedASR soft weights by window and component.
mne_denoise.viz.plot_component_summary	Plot a compact per-component summary dashboard.
mne_denoise.viz.plot_component_selector	Plot an interactive component selector for DSS or standard ZapLine.
mne_denoise.viz.plot_component_time_series	Plot stacked component time series with fixed vertical offsets.
mne_denoise.viz.plot_component_spectrogram	Plot a time-frequency power view for one component.
mne_denoise.viz.plot_component_score_curve	Plot a 1D component score curve for a fitted estimator.
mne_denoise.viz.plot_window_score_traces	Plot per-window score traces from a 2D score matrix.
mne_denoise.viz.plot_component_patterns	Plot spatial component patterns.
mne_denoise.viz.plot_component_epochs_image	Plot component activity as an epoch-by-time image.
mne_denoise.viz.plot_psd_comparison	Plot PSD comparison for original and denoised data.
mne_denoise.viz.plot_evoked_gfp_comparison	Plot GFP comparison for before/after signals.
mne_denoise.viz.plot_channel_time_course_comparison	Plot before/after channel time courses for explicit channel picks.
mne_denoise.viz.plot_power_ratio_map	Plot a topomap of preserved power ratio after denoising.
mne_denoise.viz.plot_spectrogram_comparison	Compare before/after spectrograms averaged across selected channels.
mne_denoise.viz.plot_signal_overlay	Overlay one before/after trace to inspect reconstruction quality.
mne_denoise.viz.plot_component_psd_comparison	Plot input PSD next to PSDs of selected components.
mne_denoise.viz.plot_grand_average_evokeds	Plot group-mean evoked responses with optional SEM bands.
mne_denoise.viz.plot_narrowband_score_scan	Plot score/eigenvalue profiles from a narrowband scan.
mne_denoise.viz.plot_time_frequency_mask	Visualize a time-frequency mask matrix.
mne_denoise.viz.plot_metric_bars	Plot grouped bar charts for one or more scalar metrics.
mne_denoise.viz.plot_window_count_series	Plot a per-window count or metric series.
mne_denoise.viz.plot_tradeoff_scatter	Plot a grouped x/y trade-off scatter with optional group means.
mne_denoise.viz.plot_metric_comparison	Plot one metric as grouped bars or paired subject trajectories.
mne_denoise.viz.plot_metric_slopes	Plot subject-level paired trajectories for one or more metrics.
mne_denoise.viz.plot_metric_violins	Plot violin + strip distributions with optional paired subject lines.
mne_denoise.viz.plot_null_distribution	Plot a null-distribution histogram with observed statistic and CI.
mne_denoise.viz.plot_forest	Plot per-subject point estimates with confidence intervals.
mne_denoise.viz.plot_harmonic_attenuation	Plot grouped per-harmonic attenuation bars for line-noise studies.
mne_denoise.viz.plot_metric_tradeoff_summary	Plot a two-panel metric trade-off summary.
mne_denoise.viz.plot_denoising_summary	Plot a generic denoising diagnostics dashboard.
mne_denoise.viz.plot_component_cleaning_summary	Plot a generic component-cleaning dashboard.
mne_denoise.viz.plot_signal_diagnostics_summary	Plot grouped time-domain signal diagnostics.
mne_denoise.viz.plot_condition_interaction_summary	Plot condition-by-group interaction traces.
mne_denoise.viz.plot_group_condition_interaction_summary	Plot group-wise condition interaction traces.
mne_denoise.viz.plot_endpoint_metrics_summary	Plot a generic endpoint-metric storyboard.

plot_component_selector return object

class mne_denoise.viz.ComponentSelector(*, estimator: Any, data: Any, fig: Any, state: _SelectionState, excluded: list[int])

Selection controller returned by plot_component_selector().

The controller stores the current component selection and the cached data needed by the live preview. Use excluded to inspect the selection and apply() to reconstruct the selected signal.

Notes

This class is a public return type, but direct construction is not part of the public API. Create instances with plot_component_selector().

apply(data: Any = None) → Any

Apply the current selection while preserving the input data type.

Parameters:

data (Raw | Epochs | Evoked | ndarray | None, default=None) – Data to reconstruct. If None, use the cached snapshot supplied to plot_component_selector(). Passing data explicitly computes fresh component sources for that input.

Returns:

cleaned – Reconstructed data with the same type and layout as the input. Channels not used by the fitted estimator are preserved for MNE objects.

Return type:

Raw | Epochs | Evoked | ndarray

property excluded: list[int]

Return sorted component indices excluded from the clean output.