iris_dataset_encoded

--- dataset_info: features: - name: anchor dtype: string - name: positive dtype: string - name: negative dtype: string splits: - name: train num_bytes: 36944330 num_exa