zhanghang1989 / zhanghang1989/PyTorch-Encoding

Questions about resnet101 and resnet101s

Open
#338 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
2k
Forks
448
PR merge metrics
No merged PRs in 30d

Description

We have conducted experiments with backbone resnet101 and resnet101s, respectively, leading to more than 1% mIoU gap on Pascal VOC 2012 dataset. Therefore, we want to figure out the difference between these two backbones: first, does both backbones are trained on ImageNet dataset only? Second, which one is the popular backbone used in recent researchs. Third, if we utilize resnet101s as our backbone, is it fair to compare our experiment results with other methods based on resnet101 backbone.

Contributor guide

No contributing guide indexed for this repository

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.

Research direction

The issue mentions resnet101, resnet101s, Pascal VOC 2012, and ImageNet but names no files, tests, or entry points. Start by locating the backbone definitions and training configuration, then compare their pretrained-data and evaluation settings. Done means documenting the differences and whether the comparison is methodologically fair.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
computer-vision, machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.