process.py

import json
import os
from itertools import chain
from datasets import load_dataset


os.makedirs('data', exist_ok=True)
data_list = [
    'relbert/semeval2012_relational_similarity',
    'relbert/nell_relational_similarity',
    'relbert/t_rex_relational_similarity',
    'relbert/conceptnet_relational_similarity',
]

for d in data_list:
    if d == 'relbert/t_rex_relational_similarity':
        data = load_dataset(d, "filter_unified.min_entity_4_max_predicate_10")
    else:
        data = load_dataset(d)
    for k, v in data.items():
        v = [i for i in v]
        for _v in v:
            _v['negatives'] = list(chain(*[x['positives'] for x in v if x['relation_type'] != _v['relation_type']])) + _v['negatives']
        with open(f'data/{os.path.basename(d)}.{k}.jsonl', 'w') as f:
            f.write('\n'.join([json.dumps(x) for x in v]))