Update reference scripts to fully reproduce Video Swin Transformer results

Open
#6,961 0 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
5/5
Estimated time
Over a week
Newbie friendliness
25/100
Issue type
Feature
Clarity
Needs clarification
Activity status
Stale
Tech stack
python, pytorch
Domain
computer-vision, machine-learning

Research direction

The issue names no files, tests, or entry points; begin by locating the swinTransformer3D reference scripts and the existing ported-weight implementation. Done means inflating Swin2D weights for Swin3D, supporting finetuning from specified weights, and experimentally reaching results comparable with the ported weights.

Written by the indexing model from the issue text.

Description

🚀 The feature

We currently ported the weights for swinTransformer3D. It would be nice to have references that can fully reproduce the results of paper.

Motivation, pitch

So this is a 3 step plan as discussed offline

  • As described in the paper, Use Swin2D weights weights and inflate them to make it compatible with swin3d. cc @oke-aditya
  • Modify the reference script a bit so it can do finetuning from specified weight. @YosuaMichael
  • Experiment with (2) until we get a good result comparable with the ported weight. @YosuaMichael
Alternatives

No response

Additional context

No response

Dominant language
Python
Stars
17.9k
Forks
7.3k
Avg merge
1d 15h
Merged PRs (30d)
13

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from pytorch/vision

All issues in pytorch/vision

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.