label_names.py

from typing import Dict

"""
You can update this dictionary to add more label verbalizations. 
If we find a label that is not in this dictionary, we will use the label as it is. 
If we find a label that is in this dictionary, we will replace it with the corresponding verbalization.
Verbalization can help the model to understand the label better.
"""
_label2name: Dict[str, str] = {
    "LOC": "Location",
    "PER": "Person",
    "ORG": "Organization",
    "MISC": "Miscellaneous",
    "TARGET": "Target",
    "ety": "ClinicalEntity",
    "dis": "Disability",
}


_name2label: Dict[str, str] = {v: k for k, v in _label2name.items()}


def label2name(label: str) -> str:
    if label in _label2name:
        return _label2name[label]
    else:
        return label


def name2label(name: str) -> str:
    if name in _name2label:
        return _name2label[name]
    else:
        return name