Commit History

  1. Add the unified multimodal vocabulary used by the released model

    Verified signature
    8659e37
  2. Add semRfam tokenized dataset (3,978,586 records)

    Verified signature
    e6d872e
  3. Add family -> row index lookup

    Verified signature
    a9a1da9
  4. Add family-level train/val/test split

    Verified signature
    e2dc4c7
  5. Add semRfam dataset card

    Verified signature
    a23005b
  6. initial commit

    Verified signature
    c236c35