Hugging Face's logo

Datasets:
OneThink
/
OneThinker-eval

Tasks:
Image-Text-to-Text
Video-Text-to-Text
Object Detection
Modalities:
Image
Video
Languages:
English
ArXiv:
Dataset card Data Studio Files
xet
Community
1

Enhance dataset card: Add metadata, detailed description, and sample usage

#1
by nielsr HF Staff - opened
base: refs/heads/main
←
from: refs/pr/1
Discussion Files changed
+46
-1
nielsr

This PR significantly improves the dataset card for the OneThinker-eval dataset by:

  • Adding task_categories (image-text-to-text, video-text-to-text, object-detection, image-segmentation) and language (en) to the YAML metadata block.
  • Expanding the descriptive content using information from the associated GitHub repository's README, including details about the OneThinker model and the dataset's coverage of diverse visual tasks.
  • Integrating relevant images from the GitHub README to provide better visual context.
  • Adding a 'Sample Usage' section with a command directly sourced from the GitHub repository for easy inference demonstration.
Enhance dataset card: Add metadata, detailed description, and sample usagee0bfa9b3
KaituoFeng changed pull request status to merged

Sign up or log in to comment