Skip to content

Memory issue when use STAGATE_pyG on large data #8

Description

@YingMa0107

Hi @QIFEIDKN,

Thank you for creating this great package!

I've been using STAGATE_pyG to conduct analyses across multiple sections. Each section comprises around 90,000 spatial locations, summing up to a total of about 270,000 spatial locations. During my work, I encountered an issue related to "torch.cuda.OutOfMemoryError: CUDA out of memory." Despite utilizing a GPU with more than 25GB of memory, I faced difficulties.

Considering the large size and high dimensionality of my dataset, I'm curious to know if STAGATE_pyG is intended to handle such scale. Any guidance on addressing the memory-related issue would be immensely valuable.

I truly appreciate your help and expertise!

Best

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions