Skip to content

Speed up strahler_order #335

@jefferis

Description

@jefferis

Looks like there are two places that are a bit slow

 n=neighborhood(s, 1, mode='out')

and

so_segends=so_red_nodes[match(segends, svids)]

The latter (which takes 2-3s for a 60k node neuron like mvp2, is slow because it assumes svids is unsorted (it should actually be sorted). If we could do it in one pass it would take no time at all (<1ms).

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions