Dataset · ILSVRC
Image classification via WordNet hierarchy
ILSVRC/imagenet-1k
Classic ILSVRC 2012 image dataset organized by the WordNet hierarchy for training and evaluating classifiers.
What's inside
ILSVRC 2012, known as ImageNet, is a dataset of images organized according to the WordNet hierarchy. Each WordNet concept forms a synset, with over 100k synsets in WordNet, the majority being nouns. Each synset has an average of about 1,000 images. The label language is English, and the size ranges from 1 to 10 million images.
What you can do with it
- Training or fine-tuning CNNs/ViTs for image classification
- Evaluating models on the ImageNet-1k benchmark
- Comparing the quality of image classifiers
- Tasks
- Image classification
- Languages
- English
- Size
- 1M–10M rows
- Downloads this month
- 92K
- Likes
- 1.2K
- Access
- on request on Hugging Face
Models trained on it
- KoreaPeter/ms-eff-gcvit-deepfake-b0-kodfVideo classification 73.6K
- KoreaPeter/ms-eff-gcvit-deepfake-b5-ff-plus-plusVideo classification 71.9K
- KoreaPeter/ms-eff-gcvit-deepfake-b5-celeb-df-v2Video classification 68.9K
- KoreaPeter/ms-eff-gcvit-deepfake-b0-ff-plus-plusVideo classification 63.4K
- KoreaPeter/ms-eff-gcvit-deepfake-b5-kodfVideo classification 58.8K
- KoreaPeter/ms-eff-gcvit-deepfake-b0-celeb-df-v2Video classification 55.9K
- facebook/ijepa_vith14_1kImage feature extraction 18.3K
- nvidia/MambaVision-T-1KImage classification 6.7K
How to load it
from datasets import load_dataset
ds = load_dataset("ILSVRC/imagenet-1k")