duplicated block id: 1
size: 172 cleaned lines of code
in 2 files:
 - mdr/qa/basic_tokenizer.py (16:275)
 - mdr/retrieval/utils/basic_tokenizer.py (18:277)

duplicated block id: 2
size: 73 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (37:146)
 - mdr/retrieval/single_trainer.py (37:147)

duplicated block id: 3
size: 72 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (37:144)
 - mdr/retrieval/mhop_trainer.py (38:146)

duplicated block id: 4
size: 72 cleaned lines of code
in 2 files:
 - mdr/retrieval/mhop_trainer.py (38:146)
 - mdr/retrieval/single_trainer.py (37:145)

duplicated block id: 5
size: 54 cleaned lines of code
in 2 files:
 - mdr/qa/utils.py (255:326)
 - mdr/retrieval/utils/tokenizer.py (88:158)

duplicated block id: 6
size: 49 cleaned lines of code
in 2 files:
 - mdr/qa/utils.py (41:106)
 - mdr/retrieval/utils/utils.py (21:84)

duplicated block id: 7
size: 44 cleaned lines of code
in 2 files:
 - scripts/train_mhop.py (156:206)
 - scripts/train_momentum.py (135:185)

duplicated block id: 8
size: 34 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (44:84)
 - scripts/train_mhop.py (61:100)

duplicated block id: 9
size: 34 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (44:84)
 - scripts/train_momentum.py (35:74)

duplicated block id: 10
size: 34 cleaned lines of code
in 2 files:
 - scripts/train_mhop.py (61:100)
 - scripts/train_momentum.py (35:74)

duplicated block id: 11
size: 32 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/mhop_retriever.py (56:109)
 - mdr/retrieval/models/unified_retriever.py (125:171)

duplicated block id: 12
size: 29 cleaned lines of code
in 2 files:
 - scripts/train_mhop.py (121:155)
 - scripts/train_momentum.py (96:130)

duplicated block id: 13
size: 25 cleaned lines of code
in 2 files:
 - scripts/train_mhop.py (225:254)
 - scripts/train_momentum.py (206:235)

duplicated block id: 14
size: 25 cleaned lines of code
in 2 files:
 - submitit/submitit_train.py (78:110)
 - submitit/submitit_train_qa.py (88:120)

duplicated block id: 15
size: 24 cleaned lines of code
in 2 files:
 - scripts/train_momentum.py (150:177)
 - scripts/train_qa.py (162:188)

duplicated block id: 16
size: 24 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (302:329)
 - scripts/train_qa.py (235:264)

duplicated block id: 17
size: 24 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_fever.py (57:85)
 - scripts/eval/eval_single_fever.py (46:74)

duplicated block id: 18
size: 24 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (229:253)
 - mdr/retrieval/mhop_trainer.py (223:250)

duplicated block id: 19
size: 24 cleaned lines of code
in 2 files:
 - scripts/train_mhop.py (171:198)
 - scripts/train_qa.py (162:188)

duplicated block id: 20
size: 23 cleaned lines of code
in 2 files:
 - submitit/submitit_train.py (23:57)
 - submitit/submitit_train_qa.py (22:56)

duplicated block id: 21
size: 21 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (186:211)
 - mdr/retrieval/mhop_trainer.py (181:207)

duplicated block id: 22
size: 19 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_fever.py (88:109)
 - scripts/eval/eval_single_fever.py (78:99)

duplicated block id: 23
size: 18 cleaned lines of code
in 2 files:
 - scripts/train_qa.py (264:282)
 - scripts/train_qa.py (424:441)

duplicated block id: 24
size: 18 cleaned lines of code
in 2 files:
 - mdr/retrieval/mhop_trainer.py (257:275)
 - mdr/retrieval/single_trainer.py (262:280)

duplicated block id: 25
size: 18 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (331:350)
 - scripts/train_qa.py (265:285)

duplicated block id: 26
size: 17 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (51:70)
 - mdr/retrieval/train_single.py (125:143)

duplicated block id: 27
size: 17 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (331:348)
 - scripts/train_qa.py (425:441)

duplicated block id: 28
size: 17 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (125:143)
 - scripts/train_mhop.py (68:86)

duplicated block id: 29
size: 17 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/mhop_dataset.py (14:30)
 - mdr/retrieval/data/unified_dataset.py (15:31)

duplicated block id: 30
size: 17 cleaned lines of code
in 2 files:
 - mdr/qa/utils.py (212:229)
 - mdr/retrieval/utils/tokenizer.py (48:65)

duplicated block id: 31
size: 17 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (125:143)
 - scripts/train_momentum.py (42:60)

duplicated block id: 32
size: 16 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/mhop_dataset.py (14:29)
 - mdr/retrieval/data/unified_dataset.py (254:269)

duplicated block id: 33
size: 16 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (141:160)
 - scripts/train_qa.py (154:169)

duplicated block id: 34
size: 16 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/unified_dataset.py (15:30)
 - mdr/retrieval/data/unified_dataset.py (107:122)

duplicated block id: 35
size: 16 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/fever_dataset.py (30:45)
 - mdr/retrieval/data/unified_dataset.py (15:30)

duplicated block id: 36
size: 16 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/fever_dataset.py (30:45)
 - mdr/retrieval/data/mhop_dataset.py (14:29)

duplicated block id: 37
size: 16 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/fever_dataset.py (30:45)
 - mdr/retrieval/data/unified_dataset.py (107:122)

duplicated block id: 38
size: 16 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/unified_dataset.py (15:30)
 - mdr/retrieval/data/unified_dataset.py (254:269)

duplicated block id: 39
size: 16 cleaned lines of code
in 2 files:
 - mdr/qa/qa_dataset.py (17:35)
 - mdr/retrieval/data/data_utils.py (11:29)

duplicated block id: 40
size: 16 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/fever_dataset.py (30:45)
 - mdr/retrieval/data/unified_dataset.py (254:269)

duplicated block id: 41
size: 16 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/mhop_dataset.py (14:29)
 - mdr/retrieval/data/unified_dataset.py (107:122)

duplicated block id: 42
size: 16 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (114:134)
 - scripts/train_qa.py (119:140)

duplicated block id: 43
size: 16 cleaned lines of code
in 2 files:
 - mdr/qa/qa_dataset.py (150:167)
 - mdr/qa/qa_dataset.py (302:319)

duplicated block id: 44
size: 16 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/unified_dataset.py (107:122)
 - mdr/retrieval/data/unified_dataset.py (254:269)

duplicated block id: 45
size: 15 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (311:326)
 - scripts/train_qa.py (408:423)

duplicated block id: 46
size: 15 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (115:134)
 - scripts/train_mhop.py (133:152)

duplicated block id: 47
size: 15 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (96:113)
 - scripts/train_mhop.py (114:131)

duplicated block id: 48
size: 15 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (115:134)
 - scripts/train_momentum.py (108:127)

duplicated block id: 49
size: 15 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (177:194)
 - scripts/train_mhop.py (114:131)

duplicated block id: 50
size: 15 cleaned lines of code
in 2 files:
 - scripts/train_mhop.py (133:152)
 - scripts/train_qa.py (120:140)

duplicated block id: 51
size: 15 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_fever.py (35:51)
 - scripts/eval/eval_single_fever.py (28:44)

duplicated block id: 52
size: 15 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (96:113)
 - mdr/retrieval/train_single.py (177:194)

duplicated block id: 53
size: 15 cleaned lines of code
in 2 files:
 - scripts/train_qa.py (245:260)
 - scripts/train_qa.py (408:423)

duplicated block id: 54
size: 15 cleaned lines of code
in 2 files:
 - scripts/train_momentum.py (108:127)
 - scripts/train_qa.py (120:140)

duplicated block id: 55
size: 14 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/unified_dataset.py (82:97)
 - mdr/retrieval/data/unified_dataset.py (169:184)

duplicated block id: 56
size: 14 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (94:107)
 - mdr/retrieval/data/sp_datasets.py (244:258)

duplicated block id: 57
size: 14 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (21:34)
 - mdr/retrieval/data/sp_datasets.py (94:107)

duplicated block id: 58
size: 14 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (21:34)
 - mdr/retrieval/data/sp_datasets.py (244:258)

duplicated block id: 59
size: 14 cleaned lines of code
in 2 files:
 - mdr/qa/qa_dataset.py (132:146)
 - mdr/qa/qa_dataset.py (279:298)

duplicated block id: 60
size: 13 cleaned lines of code
in 2 files:
 - mdr/retrieval/mhop_trainer.py (207:220)
 - mdr/retrieval/single_trainer.py (207:221)

duplicated block id: 61
size: 13 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/data_utils.py (182:195)
 - mdr/retrieval/data/data_utils.py (211:224)

duplicated block id: 62
size: 13 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (289:301)
 - scripts/train_qa.py (221:233)

duplicated block id: 63
size: 12 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (248:261)
 - scripts/train_momentum.py (164:177)

duplicated block id: 64
size: 12 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (248:261)
 - scripts/train_qa.py (174:188)

duplicated block id: 65
size: 12 cleaned lines of code
in 2 files:
 - mdr/qa/qa_dataset.py (169:184)
 - mdr/qa/qa_dataset.py (367:382)

duplicated block id: 66
size: 12 cleaned lines of code
in 2 files:
 - mdr/qa/basic_tokenizer.py (262:275)
 - mdr/retrieval/utils/basic_tokenizer.py (222:235)

duplicated block id: 67
size: 12 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (248:261)
 - scripts/train_mhop.py (185:198)

duplicated block id: 68
size: 12 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/mhop_retriever.py (14:28)
 - mdr/retrieval/models/retriever.py (48:61)

duplicated block id: 69
size: 12 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/fever_dataset.py (71:84)
 - mdr/retrieval/data/mhop_dataset.py (67:80)

duplicated block id: 70
size: 12 cleaned lines of code
in 2 files:
 - mdr/qa/basic_tokenizer.py (220:233)
 - mdr/qa/basic_tokenizer.py (262:275)

duplicated block id: 71
size: 12 cleaned lines of code
in 2 files:
 - mdr/retrieval/utils/basic_tokenizer.py (222:235)
 - mdr/retrieval/utils/basic_tokenizer.py (264:277)

duplicated block id: 72
size: 12 cleaned lines of code
in 2 files:
 - mdr/qa/basic_tokenizer.py (220:233)
 - mdr/retrieval/utils/basic_tokenizer.py (264:277)

duplicated block id: 73
size: 12 cleaned lines of code
in 2 files:
 - mdr/qa/utils.py (110:121)
 - mdr/retrieval/utils/utils.py (128:139)

duplicated block id: 74
size: 12 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (23:42)
 - scripts/train_qa.py (24:45)

duplicated block id: 75
size: 11 cleaned lines of code
in 2 files:
 - mdr/retrieval/utils/mhop_utils.py (67:77)
 - mdr/retrieval/utils/mhop_utils.py (86:96)

duplicated block id: 76
size: 11 cleaned lines of code
in 2 files:
 - mdr/qa/hotpot_evaluate_v1.py (13:28)
 - mdr/retrieval/utils/mhop_utils.py (37:52)

duplicated block id: 77
size: 11 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (132:143)
 - scripts/train_qa.py (61:73)

duplicated block id: 78
size: 11 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_retrieval.py (140:152)
 - scripts/eval/eval_single_fever.py (89:99)

duplicated block id: 79
size: 11 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (216:230)
 - scripts/train_momentum.py (136:148)

duplicated block id: 80
size: 11 cleaned lines of code
in 2 files:
 - scripts/train_momentum.py (49:60)
 - scripts/train_qa.py (61:73)

duplicated block id: 81
size: 11 cleaned lines of code
in 2 files:
 - scripts/train_mhop.py (75:86)
 - scripts/train_qa.py (61:73)

duplicated block id: 82
size: 11 cleaned lines of code
in 2 files:
 - mdr/retrieval/criterions.py (154:168)
 - mdr/retrieval/criterions.py (188:200)

duplicated block id: 83
size: 11 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (101:113)
 - scripts/train_qa.py (100:113)

duplicated block id: 84
size: 11 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (58:70)
 - scripts/train_qa.py (61:73)

duplicated block id: 85
size: 11 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/data_utils.py (31:46)
 - mdr/retrieval/utils/mhop_utils.py (37:52)

duplicated block id: 86
size: 11 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (182:194)
 - scripts/train_qa.py (100:113)

duplicated block id: 87
size: 11 cleaned lines of code
in 2 files:
 - mdr/retrieval/criterions.py (119:132)
 - mdr/retrieval/criterions.py (154:168)

duplicated block id: 88
size: 11 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/encode_datasets.py (18:30)
 - mdr/retrieval/data/fever_dataset.py (14:26)

duplicated block id: 89
size: 11 cleaned lines of code
in 2 files:
 - mdr/retrieval/criterions.py (119:132)
 - mdr/retrieval/criterions.py (188:200)

duplicated block id: 90
size: 11 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/mhop_dataset.py (88:103)
 - mdr/retrieval/data/unified_dataset.py (214:229)

duplicated block id: 91
size: 11 cleaned lines of code
in 2 files:
 - scripts/train_mhop.py (119:131)
 - scripts/train_qa.py (100:113)

duplicated block id: 92
size: 11 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_fever.py (99:109)
 - scripts/eval/eval_mhop_retrieval.py (140:152)

duplicated block id: 93
size: 11 cleaned lines of code
in 2 files:
 - mdr/qa/hotpot_evaluate_v1.py (13:28)
 - mdr/retrieval/data/data_utils.py (31:46)

duplicated block id: 94
size: 11 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (216:230)
 - scripts/train_mhop.py (157:169)

duplicated block id: 95
size: 10 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (213:224)
 - mdr/retrieval/single_trainer.py (209:220)

duplicated block id: 96
size: 10 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/mhop_dataset.py (105:115)
 - mdr/retrieval/data/unified_dataset.py (233:243)

duplicated block id: 97
size: 10 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (184:194)
 - scripts/train_momentum.py (96:106)

duplicated block id: 98
size: 10 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (213:224)
 - mdr/retrieval/mhop_trainer.py (209:219)

duplicated block id: 99
size: 10 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (103:113)
 - scripts/train_momentum.py (96:106)

duplicated block id: 100
size: 10 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_fever.py (83:95)
 - scripts/eval/eval_mhop_retrieval.py (79:90)

duplicated block id: 101
size: 10 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/mhop_retriever.py (14:25)
 - mdr/retrieval/models/unified_retriever.py (61:72)

duplicated block id: 102
size: 10 cleaned lines of code
in 2 files:
 - scripts/train_momentum.py (96:106)
 - scripts/train_qa.py (102:113)

duplicated block id: 103
size: 10 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/retriever.py (48:58)
 - mdr/retrieval/models/unified_retriever.py (61:72)

duplicated block id: 104
size: 9 cleaned lines of code
in 2 files:
 - scripts/train_momentum.py (138:147)
 - scripts/train_qa.py (148:157)

duplicated block id: 105
size: 9 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (218:229)
 - scripts/train_qa.py (148:157)

duplicated block id: 106
size: 9 cleaned lines of code
in 2 files:
 - mdr/qa/utils.py (29:38)
 - mdr/retrieval/utils/utils.py (10:19)

duplicated block id: 107
size: 9 cleaned lines of code
in 2 files:
 - mdr/qa/utils.py (243:256)
 - mdr/retrieval/utils/tokenizer.py (161:174)

duplicated block id: 108
size: 9 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (383:393)
 - scripts/train_qa.py (335:344)

duplicated block id: 109
size: 9 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/fever_dataset.py (71:80)
 - mdr/retrieval/data/unified_dataset.py (83:92)

duplicated block id: 110
size: 9 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/mhop_dataset.py (67:76)
 - mdr/retrieval/data/unified_dataset.py (83:92)

duplicated block id: 111
size: 9 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/mhop_dataset.py (67:76)
 - mdr/retrieval/data/unified_dataset.py (170:179)

duplicated block id: 112
size: 9 cleaned lines of code
in 2 files:
 - mdr/qa/config.py (12:23)
 - mdr/retrieval/config.py (10:21)

duplicated block id: 113
size: 9 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/fever_dataset.py (71:80)
 - mdr/retrieval/data/unified_dataset.py (170:179)

duplicated block id: 114
size: 9 cleaned lines of code
in 2 files:
 - scripts/train_mhop.py (159:168)
 - scripts/train_qa.py (148:157)

duplicated block id: 115
size: 8 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (166:173)
 - scripts/train_momentum.py (160:167)

duplicated block id: 116
size: 8 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (151:160)
 - scripts/train_mhop.py (171:179)

duplicated block id: 117
size: 8 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (151:160)
 - scripts/train_momentum.py (150:158)

duplicated block id: 118
size: 8 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_fever.py (43:50)
 - scripts/eval/eval_mhop_retrieval.py (49:56)

duplicated block id: 119
size: 8 cleaned lines of code
in 2 files:
 - mdr/retrieval/mhop_trainer.py (170:178)
 - mdr/retrieval/single_trainer.py (173:181)

duplicated block id: 120
size: 8 cleaned lines of code
in 2 files:
 - scripts/encode_corpus.py (47:54)
 - scripts/train_momentum.py (51:58)

duplicated block id: 121
size: 8 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (351:358)
 - scripts/train_qa.py (449:457)

duplicated block id: 122
size: 8 cleaned lines of code
in 2 files:
 - scripts/encode_corpus.py (47:54)
 - scripts/train_qa.py (63:71)

duplicated block id: 123
size: 8 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (237:244)
 - mdr/retrieval/single_trainer.py (246:253)

duplicated block id: 124
size: 8 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_retrieval.py (49:56)
 - scripts/eval/eval_single_fever.py (36:43)

duplicated block id: 125
size: 8 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (264:272)
 - scripts/train_mhop.py (202:210)

duplicated block id: 126
size: 8 cleaned lines of code
in 2 files:
 - mdr/retrieval/criterions.py (161:170)
 - mdr/retrieval/criterions.py (218:225)

duplicated block id: 127
size: 8 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (166:173)
 - scripts/train_mhop.py (181:188)

duplicated block id: 128
size: 8 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (170:178)
 - mdr/retrieval/single_trainer.py (173:181)

duplicated block id: 129
size: 8 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (60:68)
 - scripts/encode_corpus.py (47:54)

duplicated block id: 130
size: 8 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (170:178)
 - mdr/retrieval/mhop_trainer.py (170:178)

duplicated block id: 131
size: 8 cleaned lines of code
in 2 files:
 - mdr/retrieval/mhop_trainer.py (233:240)
 - mdr/retrieval/single_trainer.py (246:253)

duplicated block id: 132
size: 8 cleaned lines of code
in 2 files:
 - scripts/encode_corpus.py (47:54)
 - scripts/train_mhop.py (77:84)

duplicated block id: 133
size: 8 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (256:263)
 - mdr/qa/qa_trainer.py (267:274)

duplicated block id: 134
size: 8 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (166:173)
 - scripts/train_qa.py (170:177)

duplicated block id: 135
size: 8 cleaned lines of code
in 2 files:
 - mdr/retrieval/single_trainer.py (289:298)
 - mdr/retrieval/train_single.py (305:314)

duplicated block id: 136
size: 8 cleaned lines of code
in 2 files:
 - scripts/train_mhop.py (209:217)
 - scripts/train_momentum.py (190:198)

duplicated block id: 137
size: 8 cleaned lines of code
in 2 files:
 - mdr/retrieval/mhop_trainer.py (223:231)
 - mdr/retrieval/single_trainer.py (232:240)

duplicated block id: 138
size: 8 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (229:236)
 - mdr/retrieval/single_trainer.py (232:240)

duplicated block id: 139
size: 8 cleaned lines of code
in 2 files:
 - mdr/qa/config.py (42:49)
 - mdr/retrieval/config.py (40:47)

duplicated block id: 140
size: 8 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (134:141)
 - scripts/encode_corpus.py (47:54)

duplicated block id: 141
size: 7 cleaned lines of code
in 2 files:
 - mdr/qa/utils.py (232:241)
 - mdr/retrieval/utils/tokenizer.py (80:89)

duplicated block id: 142
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/retriever.py (178:184)
 - mdr/retrieval/models/unified_retriever.py (61:68)

duplicated block id: 143
size: 7 cleaned lines of code
in 2 files:
 - mdr/qa/qa_trainer.py (202:210)
 - mdr/retrieval/single_trainer.py (196:204)

duplicated block id: 144
size: 7 cleaned lines of code
in 2 files:
 - scripts/train_qa.py (293:300)
 - scripts/train_qa.py (454:461)

duplicated block id: 145
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (238:246)
 - scripts/train_mhop.py (176:183)

duplicated block id: 146
size: 7 cleaned lines of code
in 2 files:
 - scripts/train_qa.py (232:240)
 - scripts/train_qa.py (395:404)

duplicated block id: 147
size: 7 cleaned lines of code
in 2 files:
 - scripts/train_mhop.py (104:110)
 - scripts/train_momentum.py (77:83)

duplicated block id: 148
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (25:31)
 - mdr/retrieval/data/unified_dataset.py (112:118)

duplicated block id: 149
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (204:211)
 - scripts/train_mhop.py (148:155)

duplicated block id: 150
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (204:211)
 - scripts/train_momentum.py (123:130)

duplicated block id: 151
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/single_trainer.py (223:229)
 - mdr/retrieval/train_single.py (305:313)

duplicated block id: 152
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/mhop_retriever.py (14:21)
 - mdr/retrieval/models/retriever.py (178:184)

duplicated block id: 153
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (25:31)
 - mdr/retrieval/data/unified_dataset.py (259:265)

duplicated block id: 154
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (98:104)
 - mdr/retrieval/data/unified_dataset.py (20:26)

duplicated block id: 155
size: 7 cleaned lines of code
in 2 files:
 - scripts/train_momentum.py (35:42)
 - scripts/train_qa.py (47:54)

duplicated block id: 156
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (286:296)
 - scripts/train_momentum.py (206:215)

duplicated block id: 157
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/fever_dataset.py (35:41)
 - mdr/retrieval/data/sp_datasets.py (25:31)

duplicated block id: 158
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/fever_dataset.py (35:41)
 - mdr/retrieval/data/sp_datasets.py (98:104)

duplicated block id: 159
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (286:296)
 - scripts/train_mhop.py (225:234)

duplicated block id: 160
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (248:254)
 - mdr/retrieval/data/unified_dataset.py (112:118)

duplicated block id: 161
size: 7 cleaned lines of code
in 2 files:
 - scripts/train_mhop.py (61:68)
 - scripts/train_qa.py (47:54)

duplicated block id: 162
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/mhop_dataset.py (19:25)
 - mdr/retrieval/data/sp_datasets.py (98:104)

duplicated block id: 163
size: 7 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (76:82)
 - mdr/retrieval/train_single.py (145:151)

duplicated block id: 164
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (98:104)
 - mdr/retrieval/data/unified_dataset.py (112:118)

duplicated block id: 165
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/mhop_dataset.py (19:25)
 - mdr/retrieval/data/sp_datasets.py (25:31)

duplicated block id: 166
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/single_trainer.py (223:229)
 - mdr/retrieval/single_trainer.py (289:297)

duplicated block id: 167
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (74:81)
 - mdr/retrieval/data/sp_datasets.py (139:146)

duplicated block id: 168
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/fever_dataset.py (35:41)
 - mdr/retrieval/data/sp_datasets.py (248:254)

duplicated block id: 169
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (248:254)
 - mdr/retrieval/data/unified_dataset.py (20:26)

duplicated block id: 170
size: 7 cleaned lines of code
in 2 files:
 - scripts/train_qa.py (285:291)
 - scripts/train_qa.py (446:452)

duplicated block id: 171
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/retriever.py (48:54)
 - mdr/retrieval/models/retriever.py (178:184)

duplicated block id: 172
size: 7 cleaned lines of code
in 2 files:
 - mdr/qa/utils.py (145:154)
 - mdr/retrieval/utils/tokenizer.py (68:77)

duplicated block id: 173
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/mhop_dataset.py (19:25)
 - mdr/retrieval/data/sp_datasets.py (248:254)

duplicated block id: 174
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (238:246)
 - scripts/train_qa.py (166:172)

duplicated block id: 175
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (74:81)
 - mdr/retrieval/data/sp_datasets.py (204:211)

duplicated block id: 176
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (145:151)
 - scripts/train_momentum.py (66:72)

duplicated block id: 177
size: 7 cleaned lines of code
in 2 files:
 - scripts/end2end.py (49:57)
 - scripts/eval/eval_mhop_retrieval.py (44:51)

duplicated block id: 178
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (238:246)
 - scripts/train_momentum.py (155:162)

duplicated block id: 179
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (145:151)
 - scripts/train_mhop.py (92:98)

duplicated block id: 180
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (139:146)
 - mdr/retrieval/data/sp_datasets.py (204:211)

duplicated block id: 181
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (248:254)
 - mdr/retrieval/data/unified_dataset.py (259:265)

duplicated block id: 182
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (25:31)
 - mdr/retrieval/data/unified_dataset.py (20:26)

duplicated block id: 183
size: 7 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (44:51)
 - scripts/train_qa.py (47:54)

duplicated block id: 184
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (98:104)
 - mdr/retrieval/data/unified_dataset.py (259:265)

duplicated block id: 185
size: 7 cleaned lines of code
in 2 files:
 - mdr/retrieval/mhop_trainer.py (197:205)
 - mdr/retrieval/single_trainer.py (196:204)

duplicated block id: 186
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/criterions.py (193:200)
 - mdr/retrieval/criterions.py (218:223)

duplicated block id: 187
size: 6 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_retrieval.py (83:90)
 - scripts/eval/eval_single_fever.py (78:85)

duplicated block id: 188
size: 6 cleaned lines of code
in 2 files:
 - scripts/end2end.py (40:45)
 - scripts/eval/eval_mhop_retrieval.py (37:42)

duplicated block id: 189
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (154:159)
 - mdr/retrieval/data/unified_dataset.py (258:263)

duplicated block id: 190
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (154:159)
 - mdr/retrieval/data/unified_dataset.py (19:24)

duplicated block id: 191
size: 6 cleaned lines of code
in 2 files:
 - scripts/encode_corpus.py (83:89)
 - scripts/train_mhop.py (142:148)

duplicated block id: 192
size: 6 cleaned lines of code
in 2 files:
 - scripts/eval/eval_retrieval.py (89:94)
 - scripts/eval/eval_single_fever.py (36:41)

duplicated block id: 193
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/retriever.py (178:183)
 - mdr/retrieval/models/unified_retriever.py (88:94)

duplicated block id: 194
size: 6 cleaned lines of code
in 2 files:
 - mdr/qa/utils.py (130:135)
 - mdr/retrieval/utils/utils.py (132:137)

duplicated block id: 195
size: 6 cleaned lines of code
in 2 files:
 - scripts/eval/eval_retrieval.py (53:58)
 - scripts/eval/eval_single_fever.py (28:33)

duplicated block id: 196
size: 6 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (206:213)
 - scripts/train_momentum.py (208:215)

duplicated block id: 197
size: 6 cleaned lines of code
in 2 files:
 - scripts/train_mhop.py (45:59)
 - scripts/train_momentum.py (19:33)

duplicated block id: 198
size: 6 cleaned lines of code
in 2 files:
 - mdr/qa/basic_tokenizer.py (188:193)
 - mdr/retrieval/utils/basic_tokenizer.py (249:254)

duplicated block id: 199
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/criterions.py (125:132)
 - mdr/retrieval/criterions.py (218:223)

duplicated block id: 200
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (267:272)
 - mdr/retrieval/train_single.py (281:286)

duplicated block id: 201
size: 6 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_fever.py (35:40)
 - scripts/eval/eval_retrieval.py (53:58)

duplicated block id: 202
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/mhop_dataset.py (75:82)
 - mdr/retrieval/data/sp_datasets.py (321:328)

duplicated block id: 203
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/mhop_dataset.py (18:23)
 - mdr/retrieval/data/sp_datasets.py (154:159)

duplicated block id: 204
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (255:261)
 - mdr/retrieval/data/unified_dataset.py (28:33)

duplicated block id: 205
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/fever_dataset.py (34:39)
 - mdr/retrieval/data/sp_datasets.py (154:159)

duplicated block id: 206
size: 6 cleaned lines of code
in 2 files:
 - mdr/qa/utils.py (173:179)
 - mdr/retrieval/utils/tokenizer.py (39:45)

duplicated block id: 207
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/utils/basic_tokenizer.py (190:195)
 - mdr/retrieval/utils/basic_tokenizer.py (249:254)

duplicated block id: 208
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/retriever.py (48:53)
 - mdr/retrieval/models/retriever.py (72:79)

duplicated block id: 209
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/mhop_trainer.py (282:287)
 - scripts/train_momentum.py (221:226)

duplicated block id: 210
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (167:173)
 - scripts/train_momentum.py (79:86)

duplicated block id: 211
size: 6 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (124:130)
 - scripts/encode_corpus.py (83:89)

duplicated block id: 212
size: 6 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_retrieval.py (49:54)
 - scripts/eval/eval_retrieval.py (89:94)

duplicated block id: 213
size: 6 cleaned lines of code
in 2 files:
 - scripts/encode_corpus.py (83:89)
 - scripts/train_momentum.py (117:123)

duplicated block id: 214
size: 6 cleaned lines of code
in 2 files:
 - mdr/qa/config.py (55:62)
 - mdr/retrieval/config.py (69:76)

duplicated block id: 215
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (154:159)
 - mdr/retrieval/data/unified_dataset.py (111:116)

duplicated block id: 216
size: 6 cleaned lines of code
in 2 files:
 - scripts/end2end.py (40:45)
 - scripts/eval/eval_single_fever.py (28:33)

duplicated block id: 217
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/mhop_retriever.py (14:20)
 - mdr/retrieval/models/retriever.py (72:79)

duplicated block id: 218
size: 6 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_fever.py (35:40)
 - scripts/eval/eval_mhop_retrieval.py (37:42)

duplicated block id: 219
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/sp_datasets.py (109:116)
 - mdr/retrieval/data/unified_dataset.py (130:137)

duplicated block id: 220
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/retriever.py (149:154)
 - mdr/retrieval/models/unified_retriever.py (12:17)

duplicated block id: 221
size: 6 cleaned lines of code
in 2 files:
 - mdr/qa/basic_tokenizer.py (188:193)
 - mdr/qa/basic_tokenizer.py (247:252)

duplicated block id: 222
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/retriever.py (72:79)
 - mdr/retrieval/models/unified_retriever.py (88:94)

duplicated block id: 223
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/train_single.py (281:286)
 - scripts/train_mhop.py (205:210)

duplicated block id: 224
size: 6 cleaned lines of code
in 2 files:
 - mdr/qa/config.py (72:77)
 - mdr/retrieval/config.py (94:99)

duplicated block id: 225
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/retriever.py (72:79)
 - mdr/retrieval/models/retriever.py (178:183)

duplicated block id: 226
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/retriever.py (72:79)
 - mdr/retrieval/models/unified_retriever.py (61:67)

duplicated block id: 227
size: 6 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_retrieval.py (37:42)
 - scripts/eval/eval_retrieval.py (53:58)

duplicated block id: 228
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/unified_retriever.py (61:67)
 - mdr/retrieval/models/unified_retriever.py (88:94)

duplicated block id: 229
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/data/data_utils.py (80:85)
 - mdr/retrieval/data/data_utils.py (89:94)

duplicated block id: 230
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/mhop_trainer.py (282:287)
 - scripts/train_mhop.py (240:245)

duplicated block id: 231
size: 6 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_fever.py (43:48)
 - scripts/eval/eval_retrieval.py (89:94)

duplicated block id: 232
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/mhop_retriever.py (14:20)
 - mdr/retrieval/models/unified_retriever.py (88:94)

duplicated block id: 233
size: 6 cleaned lines of code
in 2 files:
 - scripts/end2end.py (40:45)
 - scripts/eval/eval_retrieval.py (53:58)

duplicated block id: 234
size: 6 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (206:213)
 - mdr/retrieval/train_single.py (288:296)

duplicated block id: 235
size: 6 cleaned lines of code
in 2 files:
 - scripts/encode_corpus.py (83:89)
 - scripts/train_qa.py (129:136)

duplicated block id: 236
size: 6 cleaned lines of code
in 2 files:
 - scripts/end2end.py (40:45)
 - scripts/eval/eval_mhop_fever.py (35:40)

duplicated block id: 237
size: 6 cleaned lines of code
in 2 files:
 - scripts/eval/eval_mhop_retrieval.py (37:42)
 - scripts/eval/eval_single_fever.py (28:33)

duplicated block id: 238
size: 6 cleaned lines of code
in 2 files:
 - mdr/qa/utils.py (114:119)
 - mdr/qa/utils.py (130:135)

duplicated block id: 239
size: 6 cleaned lines of code
in 2 files:
 - mdr/qa/basic_tokenizer.py (247:252)
 - mdr/retrieval/utils/basic_tokenizer.py (190:195)

duplicated block id: 240
size: 6 cleaned lines of code
in 2 files:
 - mdr/retrieval/models/retriever.py (48:53)
 - mdr/retrieval/models/unified_retriever.py (88:94)

duplicated block id: 241
size: 6 cleaned lines of code
in 2 files:
 - mdr/qa/train_ranker.py (206:213)
 - scripts/train_mhop.py (227:234)

duplicated block id: 242
size: 6 cleaned lines of code
in 2 files:
 - scripts/train_qa.py (322:328)
 - scripts/train_qa.py (468:474)