Skip to content
  • P
    Projects
  • G
    Groups
  • S
    Snippets
  • Help

SeetaResearch / Dragon

  • This project
    • Loading...
  • Sign in
Go to a project
  • Project
  • Repository
  • Issues 0
  • Merge Requests 0
  • Pipelines
  • Wiki
  • Snippets
  • Settings
  • Activity
  • Graph
  • Charts
  • Create a new issue
  • Jobs
  • Commits
  • Issue Boards
  • Files
  • Commits
  • Branches
  • Tags
  • Contributors
  • Graph
  • Compare
  • Charts
Switch branch/tag
  • dragon
  • torch
  • core
  • optim
  • optimizer.py
  • Ting PAN's avatar
    Fix the skipped algorithm finding in cached CUDNN convolution · a7a7e4fc
    Summary:
    This commit enforces the algorithm finding even if the backward of filter or data
    will not be executed. Otherwise, the empty algorithm will be encountered between
    two cached operation with the same arguments and input shape.
    Ting PAN committed Aug 08, 2020
    a7a7e4fc
optimizer.py 9.31 KB
BlameHistoryPermalink