Enhance dataset card: Add paper/code links, tasks, sample usage, correct license, and citations

#2
by nielsr HF Staff - opened

This PR significantly enhances the dataset card for BrainFLORA by:

  • Correcting the license in the metadata from apache-2.0 to mit as specified in the GitHub repository.
  • Adding task_categories (image-feature-extraction, image-to-image, image-to-text) and descriptive tags (neuroscience, eeg, meg, fmri, multimodal, brain-decoding, bci) for improved discoverability.
  • Including language: en metadata.
  • Linking directly to the Hugging Face paper page (https://huggingface.co/papers/2507.09747) and the GitHub repository (https://github.com/ncclab-sustech/BrainFLORA).
  • Updating the "Supported Tasks" section to clearly define "Visual Retrieval", "Visual Reconstruction", and "Visual Captioning".
  • Incorporating a "Sample Usage" section with code snippets directly from the GitHub README for environment setup, visual retrieval, reconstruction, and captioning, to help users get started easily.
  • Adding a "Citations" section with the BibTeX entries from the GitHub repository.
Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment