协作头脑风暴
brainstorming
在实现前澄清需求、探索设计选择,并形成清晰的开发方向。
surface-workflows【开发与工程】Work with Nilearn surface meshes, surface images, volume-to-surface projection, fsaverage helpers, and surface maskers while routing plotting and raw image preparation to the right Nilearn skills.
独立 GitHub 快照 · 自动分类 · 以原作者文档为准
Use this sub-skill when a task involves Nilearn cortical surface objects,
mesh/data contracts, SurfaceImage construction, volume-to-surface projection,
fsaverage meshes, or surface-specific maskers.
InMemoryMesh, FileMesh, PolyMesh,
or load_surf_mesh.PolyData, load_surf_data, or
SurfaceImage.vol_to_surf or
SurfaceImage.from_volume.SurfaceMasker, SurfaceLabelsMasker, or SurfaceMapsMasker enough to
prepare arrays for analysis pipelines.plot_surf, view_surf, report display, Plotly,
Matplotlib, or optional plotting dependencies, route to ../plotting-reporting/SKILL.md.../data-io-signal/SKILL.md.../maskers-regions/SKILL.md; keep this sub-skill focused on
surface-specific mesh/data contracts.left and/or right.SurfaceImage.(n_vertices,) for a scalar map or
(n_vertices, n_samples_or_regions) for time series, labels, or maps.fsaverage5 helper for no-network examples.SurfaceImage or surface array.references/api-reference.md summarizes public constructors, helpers, and
object relationships.references/data-formats.md describes mesh, data, hemisphere, and file-format
contracts.references/workflows.md gives recipes for minimal meshes, projection,
fsaverage, surface maskers, and plotting handoff.references/troubleshooting.md maps common surface failures to likely fixes.Run python scripts/smoke_surface.py --help for options or
python scripts/smoke_surface.py for a no-network sanity check that constructs
left/right in-memory meshes, builds a SurfaceImage, and verifies masker-style
array shapes.