mne.viz.set_3d_backend

mne.viz.set_3d_backend(backend_name, verbose=None)

Set the 3D backend for MNE.

The backend will be set as specified and operations will use that backend.

Parameters:

backend_namestr

The 3d backend to select. See Notes for the capabilities of each backend ('pyvistaqt', 'notebook' and 'jupyterlite_notebook').

Changed in version 0.24: The 'pyvista' backend was renamed 'pyvistaqt'.

verbosebool | str | int | None

Control verbosity of the logging output. If None, use the default verbosity level. See the logging documentation and mne.verbose() for details. Should only be passed as a keyword argument.

Returns:

old_backend_namestr | None

The old backend that was in use.

Notes

To use PyVista, set backend_name to pyvistaqt but the value pyvista is still supported for backward compatibility.

The jupyterlite_notebook backend is not in the table below because it is not a desktop choice: it draws with vtk.js rather than VTK, which has no WebAssembly build, and it is what the documentation’s browser notebooks run on. It covers the static 3D figures, so plot_alignment() (without channel-name labels) and plot_sparse_source_estimates() work, while mne.viz.Brain, plot_evoked_field() and snapshot_brain_montage() do not. On a desktop the other two are better in every way, so it is never selected automatically.

This table shows the capabilities of each backend (”✓” for full support, and “-” for partial support):

3D function:	pyvistaqt	notebook
plot_vector_source_estimates()	✓	✓
plot_source_estimates()	✓	✓
plot_alignment()	✓	✓
plot_sparse_source_estimates()	✓	✓
plot_evoked_field()	✓	✓
snapshot_brain_montage()	✓	✓
link_brains()	✓
Feature:
Large data	✓	✓
Opacity/transparency	✓	✓
Support geometric glyph	✓	✓
Smooth shading	✓	✓
Subplotting	✓	✓
Inline plot in Jupyter Notebook		✓
Inline plot in JupyterLab		✓
Inline plot in Google Colab
Toolbar	✓	✓