open-mmlab/mmdetection

Fix swin backbone absolute pos_embed

Open

#8,616 opened on Aug 23, 2022

View on GitHubContributor guide
 (1 comment) (1 reaction) (1 assignee)Python (9,821 forks)batch import
community help wantedgood first issuev-3.x

Repository metrics

Stars
 (32,878 stars)
PR merge metrics
 (No merged PRs in 30d)

Description

This PR needs to be migrated to dev-3.x

Originally posted by @ZwwWayne in https://github.com/open-mmlab/mmdetection/issues/8127#issuecomment-1223506509

Contributor guide

Research direction
Examine the Swin Transformer implementation in mmdetection and understand how absolute position embedding is used. Check the migration guide from dev 2.x to dev 3.x to identify required changes. Look at related issues/PRs for context.
Tech stack
pythonpytorch
Domain
backendmachine learningaicomputer vision
Issue type
Bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
Half day
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
PythonPyTorchvision transformers
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
30