Skip to content

unreasonable gpu performance of pytorch #2

@GiggleLiu

Description

@GiggleLiu

It does not even scale from intermediate scale to large scale.

  1. make sure the data are not reused (the same data to same function may not invoke reevaluation).
  2. make sure data are syncronized after evaluation.

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