mach-12 / mach-12/vit

Add additional dataloaders

Open
#1 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement good first issue
Dominant language
Python
Stars
3
Forks
0
PR merge metrics
No merged PRs in 30d

Description

There are multiple baselines used in the paper.

  1. ImageNet
  2. CIFAR-100
  3. Oxford-IIIT
  4. Pets-Oxford
  5. Flowers-102
  6. VTAB
  • Create a dataset loaders similar to load_cifar10
  • Make the traning script dynamic such that the argparse takes <model_name> as an additional argument
  • Note that the img_size in ViT model initialization at train.py is statically set as 32. You need to infer the dimensions at the data loader init. time.

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

Start with data_setup.py and the existing load_cifar10 loader, then inspect train.py and its argparse configuration. Add loaders for ImageNet, CIFAR-100, Oxford-IIIT, Pets-Oxford, Flowers-102, and VTAB; done means the training script accepts model_name dynamically and ViT receives the image dimensions inferred during loader initialization.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.