Standardisation of Dataset API split argument name
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 17.9k
- Forks
- 7.3k
- Avg merge
- 1d 15h
- Merged PRs (30d)
- 13
Description
I've noticed some naming inconsistencies across the torchvision datasets when it comes to specifying how to split the dataset (train/val/test). We currently have:
- Team
classes: LSUN - Team
image_set: SBD, VOC - Team
split: CelebA, CityScapes, ImageNet, STL10, SVHN - Team
train: CIFAR10, MNIST, USPS
The rest are unspecified - but you can effectively choose the split in them by choosing the root folder (e.g. for COCO).
Is there a reason for different naming conventions for each? If not, is there a case for standardising the argument name to one of the above so it's consistent?
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Compare the dataset constructors in lsun.py, sbd.py, voc.py, celeba.py, cityscapes.py, imagenet.py, stl10.py, svhn.py, cifar.py, mnist.py, and usps.py. First determine the preferred split argument and any compatibility implications, then standardize the affected APIs so equivalent dataset splits use one name without changing their behavior.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- computer-vision
- Issue type
- Refactor
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100