mne_rt.viz.TopomapPlot#

class mne_rt.viz.TopomapPlot(info: Info, sfreq: float, bands: dict[str, tuple[float, float]] | None = None, cmap: str = 'RdBu_r', sensors: bool = True, contours: int = 6, display_smoothing: float = 1.0, verbose=None)[source]#

Bases: _MockQMainWindow

Real-time scalp topomap showing per-band power distribution.

Displays one colour-mapped topomap per selected frequency band, updated in real-time by push(). Built on PyQt6 with a matplotlib figure embedded via FigureCanvasQTAgg.

Parameters:

infomne.Info: Channel info containing electrode positions (montage must be set).
sfreqfloat: Sampling frequency of the incoming data in Hz.
bandsdict[str, tuple[float, float]] | None, default None: Frequency bands to display as {label: (f_low, f_high)}. Defaults to δ/θ/α/β/γ standard bands.
cmapstr, default “RdBu_r”: Initial colourmap. Must be one of ["RdBu_r", "hot", "plasma", "viridis", "Reds", "RdYlBu_r"].
sensorsbool, default True: Whether to draw sensor markers on the topomap.
contoursint, default 6: Number of contour lines. Set to 0 to disable.
display_smoothingfloat, default 1.0: EMA factor applied to the per-channel band-power maps before rendering. 1.0 disables smoothing (raw per-window estimate shown directly — good for artifact monitoring); lower values progressively smooth the spatial maps across consecutive windows.
verbosebool | str | None, default None: Verbosity level. See set_log_level().

See also

mne_rt.viz.NFPlot: Scrolling NF signal display.
mne_rt.viz.BrainPlot: 3D brain activation display.
mne_rt.RTStream.record_main: Main NF loop that drives all plots.

Notes

The control panel (right sidebar) provides:

Playback — pause/resume, screenshot.
Display — colormap selector, contours, sensor toggle, colorbar mode.
Bands — per-band visibility checkboxes. Toggling rebuilds the figure layout automatically.

Band power is estimated via FFT on each incoming window (Hanning window applied to reduce spectral leakage).

Examples

Minimal offline usage:

>>> from mne import create_info
>>> import numpy as np
>>> app = QApplication([])
>>> info = create_info(["Fp1", "Fz", "Cz", "Oz"], sfreq=256, ch_types="eeg")
>>> plot = TopomapPlot(info, sfreq=256)
>>> plot.show()
>>> data = np.random.randn(4, 256)   # 1 s window
>>> plot.push(data)
>>> app.exec()

Added in version 1.0.0.

__init__(info: Info, sfreq: float, bands: dict[str, tuple[float, float]] | None = None, cmap: str = 'RdBu_r', sensors: bool = True, contours: int = 6, display_smoothing: float = 1.0, verbose=None) → None[source]#

Methods

`__init__`(info, sfreq[, bands, cmap, ...])
`push`(data)	Update all visible topomaps from a new raw data window.

push(data: ndarray) → None[source]#

Update all visible topomaps from a new raw data window.

Called at ~5 fps by the Qt pump timer inside record_main().

Parameters:

datandarray of shape (n_channels, n_times): Raw EEG/MEG data for the current analysis window.

Notes

The call is a no-op when paused (⏸ button pressed). Band power is estimated via FFT and the topomaps are redrawn via mne.viz.plot_topomap().