Skip to content

Run GCN end2end test for different number of layers(2,4,6,8) to show effect of fusion for more than one layer. #56

Description

@mmsalehid
  • Put separate measurements for backward pass and forward pass in the python training code.
  • Run PyTorch GCN training and DGL GCN Training for different number of layers and with or without dropout.
  • Extract the timing for different parts with different settings.

Activity

  1. self-assigned this
    on Mar 13, 2024
  2. changed the title [-]Run GCN end to end for different number of layers(2,4,6,8) to show effect of fusion for more than one layer.[/-] [+]Run GeMM-SpMM for different number of layers(2,4,6,8) to show effect of fusion for more than one layer.[/+] on Mar 14, 2024
  3. changed the title [-]Run GeMM-SpMM for different number of layers(2,4,6,8) to show effect of fusion for more than one layer.[/-] [+]Run GCN end2end test for different number of layers(2,4,6,8) to show effect of fusion for more than one layer.[/+] on Mar 14, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

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