Skip to content

feat(dreamzero): change default video backend to torchcodec#1159

Open
thereAreDemonsNearby wants to merge 1 commit into
RLinf:mainfrom
thereAreDemonsNearby:feature/faster_video_decode
Open

feat(dreamzero): change default video backend to torchcodec#1159
thereAreDemonsNearby wants to merge 1 commit into
RLinf:mainfrom
thereAreDemonsNearby:feature/faster_video_decode

Conversation

@thereAreDemonsNearby
Copy link
Copy Markdown
Collaborator

Description

change default video backend to torchcodec

Motivation and Context

torchcodec is faster than the old default pyav

How has this been tested?

Spikes are far less (yellow is old, blue is new):
img_v3_0211m_e2e45a3e-7992-4edd-ae73-05c9b0bb02bg

Loss curves are basically the same:
img_v3_0211m_d6250362-d1c5-451d-8749-7e2ab24f4f6g

Additional information (optional, e.g., figures and logs):

Types of changes

  • Bug fix (non-breaking change which fixes an issue)
  • New feature (non-breaking change which adds functionality)
  • Documentation update (Document-only update)
  • Breaking change (fix or feature that would cause existing functionality to not work as expected)

Checklist:

  • My code follows the code style of this project.
  • My change requires a change to the documentation.
  • I have updated the documentation accordingly.
  • I have added tests to cover my changes.
  • All new and existing tests passed.

Signed-off-by: shiletong <shiletong@infini-ai.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants