model update

generate_reference_files.py

@@ -1,16 +0,0 @@
-import os
-from glob import glob
-from datasets import load_dataset
-
-os.makedirs('./reference_files', exist_ok=True)
-
-
-for split in ['validation', 'test']:
-    dataset = load_dataset('lmqg/qag_squad', split=split, download_mode='force_redownload')
-    for data in ['questions_answers', 'paragraph']:
-        with open('./reference_files/{}-{}.txt'.format(data, split), 'w') as f:
-            tmp_data = dataset[data]
-            f.write('\n'.join([i.replace('\n', '') for i in tmp_data]))
-    length = [len(open(i).read().split('\n')) for i in glob(f'reference_files/*{split}.txt')]
-    assert len(list(set(length))) == 1, length
-    assert length[0] == len(dataset), f"{length[0]} != {len(dataset)}"