This dataset contains the COCO-2014 images and captions converted to WebDataset (WDS) format, using the Karpathy & Li (2015) dataset split for image captioning tasks.
Total Samples: 123,287 images with 5 reference captions each
Total Size: ~19 GB
Format: WebDataset (.tar shards)
Shard Size: 1,000 samples per tar file
License: CC-BY 4.0
Language: English
COCO-2014-WDS/
├── train/ (113… See the full description on the dataset page:
https://huggingface.co/datasets/undefined443/coco-karpathy-wds.
