Skip to content

[FEA] GPU accelerated Phate? #679

@srivarra

Description

@srivarra

Is your feature request related to a problem? Please describe.

I'd love for a GPU accelerated implementation of PHATE as many of my colleagues arn't able to scale it well and have it run in a reasonable amount of time.

Describe the solution you'd like

I'd like a CuPy implementation of PHATE (something like rsc.tl.phate) that can be ran out of core via dask and scale across multiple GPUs on a node in an HPC.

Is there a CPU based implementation

Implementation here: PHATE docs.

If this is in scope for rapids-singlecell, I'd love to contribute this feature.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels
    No fields configured for Enhancement.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions