Skip to content

T34: Pretrained fine-tuning regime on Imagewoof #342

Description

@mateuszwalo

Track: research · Level: core · Effort: ~5h (2h GPU) · Depends on: #335 (T20)

Why this matters

Most real users fine-tune a pretrained backbone rather than train from scratch. Showing BNNR still helps in that regime makes the method relevant to how people actually work, not just to a research setup.

Steps

  1. Get maintainer sign-off.
  2. Run the Imagewoof matrix with the pretrained option (check run_grand_benchmark.py --help for the regime flag).
  3. Summarize with summarize_grand and write findings.md.

Done when

Pretrained Imagewoof JSON; table; findings.md.


How to take this: comment "taking this" and wait to be assigned. Branch t34-short-desc from upstream/main, and put Closes #<this issue number> in your PR. Full workflow: the Cohort Handbook (pinned in Discord).

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions