---
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:17247827
- loss:RZTKMatryoshka2dLoss
base_model: intfloat/multilingual-e5-base
widget:
- source_sentence: 'query: толокно'
sentences:
- 'passage: Видеорегистратор DVR G30'
- 'passage: Толокно вівсяне 500 г ТМ Козуб Продукт'
- 'passage: Набор открытка + лист наклеек Poof Poof "Only love" miss you'
- source_sentence: 'query: корпус'
sentences:
- 'passage: Корпуси Thermaltake Гарантія 12 місяців Форм-фактор материнської плати
ATX Тип корпусу Miditower Вид Геймерський Колір Білий Кількість вантажних місць
1 Країна реєстрації бренда Китай (Тайвань) Країна-виробник товару Китай Кількість
внутрішніх відсіків 3.5" 3 теги Круті теги Кастомні теги Прозорі Спосіб встановлення
Вертикальний Розташування БЖ Нижнє Функціонал і роз''єми лицьової панелі Вихід
для навушників Функціонал і роз''єми лицьової панелі Вхід для мікрофона Функціонал
і роз''єми лицьової панелі 2 х USB 3.0 Розташування роз''ємів лицьової панелі
Зверху Матеріал Загартоване скло Матеріал Сталь'
- 'passage: Мягкие игрушки Цвет Желтый Высота игрушки, см 22 теги популярные Детский
возраст 3 года'
- 'passage: Корпус Сrown CMC-SM159 500 Вт Black-Grey (CMC-SM159 500W)'
- source_sentence: 'query: пижама для мальчика'
sentences:
- 'passage: Батарейка Enerlight Mega Power D/RL20 2 шт (90200102)'
- 'passage: Піжами для хлопчиків Витуся Колір Світло-блакитний Зріст 98 см'
- 'passage: Корпуси QUBE Гарантія 36 місяців Форм-фактор материнської плати ATX
Тип корпусу Fulltower Блок живлення Без блока живлення Колір Чорний + Помаранчевий
Максимальна висота кулера ЦП 170 теги Круті Спосіб встановлення Вертикальний Розташування
БЖ Нижнє Матеріал Метал'
- source_sentence: 'query: кисточки для макияжа'
sentences:
- 'passage: Кисті для макіяжу Kylie 12 шт набір кистей пензлика 12 шт Білі'
- 'passage: Корпус 1stPlayer A1-450PLS 450W Black-12cm, 2хUSB2.0, 1хUSB3.0'
- 'passage: Набор открытка + лист наклеек Poof Poof "Only love" miss you'
- source_sentence: 'query: омивач зимовий'
sentences:
- 'passage: Набор офисной бумаги цветной Magnat Color А4 80г/м.кв 4цветовХ50листов
NEON'
- 'passage: Атомайзери для парфумів Країна-виробник товару Китай Об''єм 5 мл Колір
Чорний'
- 'passage: Рідини для склоомивачів Тайга Консистенція Рідина Країна реєстрації
бренда Україна Країна-виробник товару Україна Марка автомобіля Всі марки Сезон
Зима'
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
- dot_accuracy@1
- dot_accuracy@3
- dot_accuracy@5
- dot_accuracy@10
- dot_precision@1
- dot_precision@3
- dot_precision@5
- dot_precision@10
- dot_recall@1
- dot_recall@3
- dot_recall@5
- dot_recall@10
- dot_ndcg@10
- dot_mrr@10
- dot_map@100
- dot_ndcg@1
- dot_mrr@1
model-index:
- name: SentenceTransformer based on intfloat/multilingual-e5-base
results:
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: bm full
type: bm-full
metrics:
- type: dot_accuracy@1
value: 0.4688605803255485
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.6298655343241331
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.7222222222222222
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.8152866242038217
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.4688605803255485
name: Dot Precision@1
- type: dot_precision@3
value: 0.4742863882991272
name: Dot Precision@3
- type: dot_precision@5
value: 0.48400566171266807
name: Dot Precision@5
- type: dot_precision@10
value: 0.48927813163481954
name: Dot Precision@10
- type: dot_recall@1
value: 0.011148491409474373
name: Dot Recall@1
- type: dot_recall@3
value: 0.0336986651180224
name: Dot Recall@3
- type: dot_recall@5
value: 0.0574639763212004
name: Dot Recall@5
- type: dot_recall@10
value: 0.11438221839538937
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.4853995206687327
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.5667621519001553
name: Dot Mrr@10
- type: dot_map@100
value: 0.36319946056506935
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: core uk title
type: core-uk-title
metrics:
- type: dot_accuracy@1
value: 0.6314168377823408
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.8501026694045175
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.9188911704312115
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.9722792607802875
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.6314168377823408
name: Dot Precision@1
- type: dot_precision@3
value: 0.6389459274469542
name: Dot Precision@3
- type: dot_precision@5
value: 0.619917864476386
name: Dot Precision@5
- type: dot_precision@10
value: 0.5194045174537988
name: Dot Precision@10
- type: dot_recall@1
value: 0.06422498114278523
name: Dot Recall@1
- type: dot_recall@3
value: 0.19465571989578512
name: Dot Recall@3
- type: dot_recall@5
value: 0.31310863697776364
name: Dot Recall@5
- type: dot_recall@10
value: 0.5138760113039917
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.5824172083713806
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.7521805025911807
name: Dot Mrr@10
- type: dot_map@100
value: 0.5633455872213496
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: core ru title
type: core-ru-title
metrics:
- type: dot_accuracy@1
value: 0.6283367556468172
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.8562628336755647
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.9240246406570842
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.973305954825462
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.6283367556468172
name: Dot Precision@1
- type: dot_precision@3
value: 0.6468172484599589
name: Dot Precision@3
- type: dot_precision@5
value: 0.6303901437371663
name: Dot Precision@5
- type: dot_precision@10
value: 0.526694045174538
name: Dot Precision@10
- type: dot_recall@1
value: 0.0642355037964411
name: Dot Recall@1
- type: dot_recall@3
value: 0.196867872903626
name: Dot Recall@3
- type: dot_recall@5
value: 0.31877735224606824
name: Dot Recall@5
- type: dot_recall@10
value: 0.5205272597280992
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.5898897836994542
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.7534817965516121
name: Dot Mrr@10
- type: dot_map@100
value: 0.5705926428712759
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: core uk options
type: core-uk-options
metrics:
- type: dot_accuracy@1
value: 0.4928131416837782
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.7453798767967146
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.8449691991786448
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.9353182751540041
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.4928131416837782
name: Dot Precision@1
- type: dot_precision@3
value: 0.5068446269678302
name: Dot Precision@3
- type: dot_precision@5
value: 0.5071868583162218
name: Dot Precision@5
- type: dot_precision@10
value: 0.4586242299794661
name: Dot Precision@10
- type: dot_recall@1
value: 0.04944135228891842
name: Dot Recall@1
- type: dot_recall@3
value: 0.1526358179003426
name: Dot Recall@3
- type: dot_recall@5
value: 0.25373048761441075
name: Dot Recall@5
- type: dot_recall@10
value: 0.45235075750838527
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.49534030959419667
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.6405511554382188
name: Dot Mrr@10
- type: dot_map@100
value: 0.49970085786624235
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: core ru options
type: core-ru-options
metrics:
- type: dot_accuracy@1
value: 0.49691991786447637
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.7700205338809035
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.8470225872689938
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.9312114989733059
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.49691991786447637
name: Dot Precision@1
- type: dot_precision@3
value: 0.5109514031485284
name: Dot Precision@3
- type: dot_precision@5
value: 0.506776180698152
name: Dot Precision@5
- type: dot_precision@10
value: 0.4619096509240247
name: Dot Precision@10
- type: dot_recall@1
value: 0.049396814064195396
name: Dot Recall@1
- type: dot_recall@3
value: 0.15343729595148847
name: Dot Recall@3
- type: dot_recall@5
value: 0.2530940085166139
name: Dot Recall@5
- type: dot_recall@10
value: 0.4551132844186051
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.49867060052138473
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.6456927740295298
name: Dot Mrr@10
- type: dot_map@100
value: 0.4993779830024849
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: options uk title
type: options-uk-title
metrics:
- type: dot_accuracy@1
value: 0.779510022271715
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.9242761692650334
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.9621380846325167
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.9888641425389755
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.779510022271715
name: Dot Precision@1
- type: dot_precision@3
value: 0.755011135857461
name: Dot Precision@3
- type: dot_precision@5
value: 0.7167037861915367
name: Dot Precision@5
- type: dot_precision@10
value: 0.5592427616926503
name: Dot Precision@10
- type: dot_recall@1
value: 0.11936675054715144
name: Dot Recall@1
- type: dot_recall@3
value: 0.34548403217668255
name: Dot Recall@3
- type: dot_recall@5
value: 0.5402662482283863
name: Dot Recall@5
- type: dot_recall@10
value: 0.7986739009567517
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.7729516881303515
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.8563271113939265
name: Dot Mrr@10
- type: dot_map@100
value: 0.7473933597845402
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: options ru title
type: options-ru-title
metrics:
- type: dot_accuracy@1
value: 0.779510022271715
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.9287305122494433
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.9710467706013363
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.9955456570155902
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.779510022271715
name: Dot Precision@1
- type: dot_precision@3
value: 0.7572383073496659
name: Dot Precision@3
- type: dot_precision@5
value: 0.7220489977728285
name: Dot Precision@5
- type: dot_precision@10
value: 0.555902004454343
name: Dot Precision@10
- type: dot_recall@1
value: 0.11926776514749789
name: Dot Recall@1
- type: dot_recall@3
value: 0.3470554253961826
name: Dot Recall@3
- type: dot_recall@5
value: 0.5424129940833727
name: Dot Recall@5
- type: dot_recall@10
value: 0.7947127173741874
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.7720861199268557
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.8601848905857812
name: Dot Mrr@10
- type: dot_map@100
value: 0.7485857122000162
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: options uk options
type: options-uk-options
metrics:
- type: dot_accuracy@1
value: 0.6369710467706013
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.8485523385300668
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.910913140311804
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.9599109131403119
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.6369710467706013
name: Dot Precision@1
- type: dot_precision@3
value: 0.6302895322939867
name: Dot Precision@3
- type: dot_precision@5
value: 0.6160356347438753
name: Dot Precision@5
- type: dot_precision@10
value: 0.5042316258351893
name: Dot Precision@10
- type: dot_recall@1
value: 0.09330619269372055
name: Dot Recall@1
- type: dot_recall@3
value: 0.27903919886102513
name: Dot Recall@3
- type: dot_recall@5
value: 0.45307867089604287
name: Dot Recall@5
- type: dot_recall@10
value: 0.7070577714787069
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.6688855091280834
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.7514184961289637
name: Dot Mrr@10
- type: dot_map@100
value: 0.6485816948624684
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: options ru options
type: options-ru-options
metrics:
- type: dot_accuracy@1
value: 0.6369710467706013
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.844097995545657
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.8930957683741648
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.9532293986636972
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.6369710467706013
name: Dot Precision@1
- type: dot_precision@3
value: 0.6362286562731996
name: Dot Precision@3
- type: dot_precision@5
value: 0.6138084632516704
name: Dot Precision@5
- type: dot_precision@10
value: 0.49888641425389757
name: Dot Precision@10
- type: dot_recall@1
value: 0.0944295002201461
name: Dot Recall@1
- type: dot_recall@3
value: 0.2840927570326234
name: Dot Recall@3
- type: dot_recall@5
value: 0.4497944767432518
name: Dot Recall@5
- type: dot_recall@10
value: 0.7009835067184734
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.6652750740822607
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.7476314207940042
name: Dot Mrr@10
- type: dot_map@100
value: 0.6484652109747833
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: rusisms uk title
type: rusisms-uk-title
metrics:
- type: dot_accuracy@1
value: 0.6412698412698413
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.7904761904761904
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.8412698412698413
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.8888888888888888
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.6412698412698413
name: Dot Precision@1
- type: dot_precision@3
value: 0.6507936507936508
name: Dot Precision@3
- type: dot_precision@5
value: 0.6406349206349207
name: Dot Precision@5
- type: dot_precision@10
value: 0.6123809523809524
name: Dot Precision@10
- type: dot_recall@1
value: 0.040916125312104056
name: Dot Recall@1
- type: dot_recall@3
value: 0.12029766150988147
name: Dot Recall@3
- type: dot_recall@5
value: 0.1766554074779815
name: Dot Recall@5
- type: dot_recall@10
value: 0.30149841054052495
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.6530771088562558
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.7262358276643992
name: Dot Mrr@10
- type: dot_map@100
value: 0.5739194742316258
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: rusisms ru title
type: rusisms-ru-title
metrics:
- type: dot_accuracy@1
value: 0.6444444444444445
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.7714285714285715
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.8222222222222222
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.8825396825396825
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.6444444444444445
name: Dot Precision@1
- type: dot_precision@3
value: 0.6529100529100529
name: Dot Precision@3
- type: dot_precision@5
value: 0.6444444444444445
name: Dot Precision@5
- type: dot_precision@10
value: 0.6174603174603175
name: Dot Precision@10
- type: dot_recall@1
value: 0.04345252136983852
name: Dot Recall@1
- type: dot_recall@3
value: 0.117225830891467
name: Dot Recall@3
- type: dot_recall@5
value: 0.17312874488116914
name: Dot Recall@5
- type: dot_recall@10
value: 0.2975741221948891
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.6565427517594488
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.7232627865961199
name: Dot Mrr@10
- type: dot_map@100
value: 0.5819623451494035
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: rusisms uk options
type: rusisms-uk-options
metrics:
- type: dot_accuracy@1
value: 0.5047619047619047
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.638095238095238
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.6952380952380952
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.8031746031746032
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.5047619047619047
name: Dot Precision@1
- type: dot_precision@3
value: 0.5164021164021164
name: Dot Precision@3
- type: dot_precision@5
value: 0.5161904761904763
name: Dot Precision@5
- type: dot_precision@10
value: 0.5107936507936507
name: Dot Precision@10
- type: dot_recall@1
value: 0.031944436826188825
name: Dot Recall@1
- type: dot_recall@3
value: 0.08279504216625777
name: Dot Recall@3
- type: dot_recall@5
value: 0.12790491040181537
name: Dot Recall@5
- type: dot_recall@10
value: 0.24261713123599732
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.5348137564536981
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.5903086419753085
name: Dot Mrr@10
- type: dot_map@100
value: 0.49149267540077934
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: rusisms ru options
type: rusisms-ru-options
metrics:
- type: dot_accuracy@1
value: 0.5079365079365079
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.6507936507936508
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.7206349206349206
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.7936507936507936
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.5079365079365079
name: Dot Precision@1
- type: dot_precision@3
value: 0.5248677248677248
name: Dot Precision@3
- type: dot_precision@5
value: 0.5219047619047619
name: Dot Precision@5
- type: dot_precision@10
value: 0.5209523809523811
name: Dot Precision@10
- type: dot_recall@1
value: 0.032188392229080466
name: Dot Recall@1
- type: dot_recall@3
value: 0.08464291377132598
name: Dot Recall@3
- type: dot_recall@5
value: 0.12889238376608458
name: Dot Recall@5
- type: dot_recall@10
value: 0.2460953092989818
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.5427041762757129
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.5953653313177122
name: Dot Mrr@10
- type: dot_map@100
value: 0.49522979085271474
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: rusisms corrected uk title
type: rusisms_corrected-uk-title
metrics:
- type: dot_accuracy@1
value: 0.7151898734177216
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.8259493670886076
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.8892405063291139
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.9177215189873418
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.7151898734177216
name: Dot Precision@1
- type: dot_precision@3
value: 0.7078059071729957
name: Dot Precision@3
- type: dot_precision@5
value: 0.709493670886076
name: Dot Precision@5
- type: dot_precision@10
value: 0.6699367088607595
name: Dot Precision@10
- type: dot_recall@1
value: 0.04989155154757441
name: Dot Recall@1
- type: dot_recall@3
value: 0.12806316976411466
name: Dot Recall@3
- type: dot_recall@5
value: 0.19810605504728512
name: Dot Recall@5
- type: dot_recall@10
value: 0.3338919149499279
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.7184486440976392
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.786057112718505
name: Dot Mrr@10
- type: dot_map@100
value: 0.6363110580408179
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: rusisms corrected ru title
type: rusisms_corrected-ru-title
metrics:
- type: dot_accuracy@1
value: 0.7025316455696202
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.8164556962025317
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.8734177215189873
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.9145569620253164
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.7025316455696202
name: Dot Precision@1
- type: dot_precision@3
value: 0.7067510548523206
name: Dot Precision@3
- type: dot_precision@5
value: 0.7
name: Dot Precision@5
- type: dot_precision@10
value: 0.6715189873417721
name: Dot Precision@10
- type: dot_recall@1
value: 0.0481861847768873
name: Dot Recall@1
- type: dot_recall@3
value: 0.12691328806666166
name: Dot Recall@3
- type: dot_recall@5
value: 0.1961403720941317
name: Dot Recall@5
- type: dot_recall@10
value: 0.3375990992674748
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.7170183596007423
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.7751092525617841
name: Dot Mrr@10
- type: dot_map@100
value: 0.6388651896818722
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: rusisms corrected uk options
type: rusisms_corrected-uk-options
metrics:
- type: dot_accuracy@1
value: 0.5664556962025317
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.7025316455696202
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.7689873417721519
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.8544303797468354
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.5664556962025317
name: Dot Precision@1
- type: dot_precision@3
value: 0.5875527426160337
name: Dot Precision@3
- type: dot_precision@5
value: 0.590506329113924
name: Dot Precision@5
- type: dot_precision@10
value: 0.5768987341772153
name: Dot Precision@10
- type: dot_recall@1
value: 0.03655980998007077
name: Dot Recall@1
- type: dot_recall@3
value: 0.09868763166111327
name: Dot Recall@3
- type: dot_recall@5
value: 0.15664912559151947
name: Dot Recall@5
- type: dot_recall@10
value: 0.2920921556547295
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.607554678674386
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.6545320976491863
name: Dot Mrr@10
- type: dot_map@100
value: 0.5617770075337222
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: rusisms corrected ru options
type: rusisms_corrected-ru-options
metrics:
- type: dot_accuracy@1
value: 0.5759493670886076
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.7088607594936709
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.7911392405063291
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.8734177215189873
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.5759493670886076
name: Dot Precision@1
- type: dot_precision@3
value: 0.5780590717299579
name: Dot Precision@3
- type: dot_precision@5
value: 0.5886075949367089
name: Dot Precision@5
- type: dot_precision@10
value: 0.5854430379746836
name: Dot Precision@10
- type: dot_recall@1
value: 0.0372137314078135
name: Dot Recall@1
- type: dot_recall@3
value: 0.09554487551202463
name: Dot Recall@3
- type: dot_recall@5
value: 0.15170114399950857
name: Dot Recall@5
- type: dot_recall@10
value: 0.2975626314435191
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.6138258669469168
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.6650843881856539
name: Dot Mrr@10
- type: dot_map@100
value: 0.5642621603475558
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: core typos uk title
type: core_typos-uk-title
metrics:
- type: dot_accuracy@1
value: 0.5431211498973306
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.7741273100616016
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.8490759753593429
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.9034907597535934
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.5431211498973306
name: Dot Precision@1
- type: dot_precision@3
value: 0.5379876796714579
name: Dot Precision@3
- type: dot_precision@5
value: 0.5215605749486653
name: Dot Precision@5
- type: dot_precision@10
value: 0.44271047227926075
name: Dot Precision@10
- type: dot_recall@1
value: 0.05559195278892706
name: Dot Recall@1
- type: dot_recall@3
value: 0.1639580744107257
name: Dot Recall@3
- type: dot_recall@5
value: 0.2637510372155257
name: Dot Recall@5
- type: dot_recall@10
value: 0.439356948127633
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.49609977907989355
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.6681244092435059
name: Dot Mrr@10
- type: dot_map@100
value: 0.4701798686376467
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: core typos ru title
type: core_typos-ru-title
metrics:
- type: dot_accuracy@1
value: 0.5544147843942505
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.7720739219712526
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.8459958932238193
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.9086242299794661
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.5544147843942505
name: Dot Precision@1
- type: dot_precision@3
value: 0.5458590006844627
name: Dot Precision@3
- type: dot_precision@5
value: 0.5332648870636552
name: Dot Precision@5
- type: dot_precision@10
value: 0.44825462012320333
name: Dot Precision@10
- type: dot_recall@1
value: 0.0567222289494001
name: Dot Recall@1
- type: dot_recall@3
value: 0.16720128524694292
name: Dot Recall@3
- type: dot_recall@5
value: 0.27062845238802474
name: Dot Recall@5
- type: dot_recall@10
value: 0.4452366938264994
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.5033845006448656
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.6771202046869397
name: Dot Mrr@10
- type: dot_map@100
value: 0.4771638964943432
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: core typos uk options
type: core_typos-uk-options
metrics:
- type: dot_accuracy@1
value: 0.4004106776180698
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.6365503080082136
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.7361396303901437
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.8459958932238193
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.4004106776180698
name: Dot Precision@1
- type: dot_precision@3
value: 0.4144421629021219
name: Dot Precision@3
- type: dot_precision@5
value: 0.4106776180698152
name: Dot Precision@5
- type: dot_precision@10
value: 0.373305954825462
name: Dot Precision@10
- type: dot_recall@1
value: 0.03972231935905085
name: Dot Recall@1
- type: dot_recall@3
value: 0.1253059683569105
name: Dot Recall@3
- type: dot_recall@5
value: 0.20604960082327517
name: Dot Recall@5
- type: dot_recall@10
value: 0.3702481690780398
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.40438785285575113
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.5403185196049667
name: Dot Mrr@10
- type: dot_map@100
value: 0.4013217513460408
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: core typos ru options
type: core_typos-ru-options
metrics:
- type: dot_accuracy@1
value: 0.40451745379876797
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.6457905544147844
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.7433264887063655
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.8480492813141683
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.40451745379876797
name: Dot Precision@1
- type: dot_precision@3
value: 0.42128678986995205
name: Dot Precision@3
- type: dot_precision@5
value: 0.42073921971252576
name: Dot Precision@5
- type: dot_precision@10
value: 0.3737166324435318
name: Dot Precision@10
- type: dot_recall@1
value: 0.04056242195283746
name: Dot Recall@1
- type: dot_recall@3
value: 0.12692474924855043
name: Dot Recall@3
- type: dot_recall@5
value: 0.2108604051630081
name: Dot Recall@5
- type: dot_recall@10
value: 0.3708740498539991
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.4067414113253185
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.5483214367197938
name: Dot Mrr@10
- type: dot_map@100
value: 0.40227123839479567
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: 'bm full matryoshka dim 768 '
type: bm-full--matryoshka_dim-768--
metrics:
- type: dot_accuracy@1
value: 0.4688605803255485
name: Dot Accuracy@1
- type: dot_precision@1
value: 0.4688605803255485
name: Dot Precision@1
- type: dot_recall@1
value: 0.011148491409474373
name: Dot Recall@1
- type: dot_ndcg@1
value: 0.4688605803255485
name: Dot Ndcg@1
- type: dot_mrr@1
value: 0.4688605803255485
name: Dot Mrr@1
- type: dot_map@100
value: 0.36319946056506935
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: 'bm full matryoshka dim 512 '
type: bm-full--matryoshka_dim-512--
metrics:
- type: dot_accuracy@1
value: 0.4692144373673036
name: Dot Accuracy@1
- type: dot_precision@1
value: 0.4692144373673036
name: Dot Precision@1
- type: dot_recall@1
value: 0.011025230024960227
name: Dot Recall@1
- type: dot_ndcg@1
value: 0.4692144373673036
name: Dot Ndcg@1
- type: dot_mrr@1
value: 0.4692144373673036
name: Dot Mrr@1
- type: dot_map@100
value: 0.3612995014580172
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: 'bm full matryoshka dim 256 '
type: bm-full--matryoshka_dim-256--
metrics:
- type: dot_accuracy@1
value: 0.4731068648266101
name: Dot Accuracy@1
- type: dot_precision@1
value: 0.4731068648266101
name: Dot Precision@1
- type: dot_recall@1
value: 0.011166576278060827
name: Dot Recall@1
- type: dot_ndcg@1
value: 0.4731068648266101
name: Dot Ndcg@1
- type: dot_mrr@1
value: 0.4731068648266101
name: Dot Mrr@1
- type: dot_map@100
value: 0.35359376379148333
name: Dot Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: 'bm full matryoshka dim 128 '
type: bm-full--matryoshka_dim-128--
metrics:
- type: dot_accuracy@1
value: 0.46602972399150744
name: Dot Accuracy@1
- type: dot_precision@1
value: 0.46602972399150744
name: Dot Precision@1
- type: dot_recall@1
value: 0.010987678469890481
name: Dot Recall@1
- type: dot_ndcg@1
value: 0.46602972399150744
name: Dot Ndcg@1
- type: dot_mrr@1
value: 0.46602972399150744
name: Dot Mrr@1
- type: dot_map@100
value: 0.33458221563344687
name: Dot Map@100
---
# SentenceTransformer based on intfloat/multilingual-e5-base
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/multilingual-e5-base](https://huggingface.co/intfloat/multilingual-e5-base) on the rozetka_positive_pairs dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [intfloat/multilingual-e5-base](https://huggingface.co/intfloat/multilingual-e5-base)
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 768 dimensions
- **Similarity Function:** Dot Product
- **Training Dataset:**
- rozetka_positive_pairs
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
RZTKSentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("yklymchuk-rztk/multilingual-e5-base-matryoshka2d-mnr-5")
# Run inference
sentences = [
'query: омивач зимовий',
'passage: Рідини для склоомивачів Тайга Консистенція Рідина Країна реєстрації бренда Україна Країна-виробник товару Україна Марка автомобіля Всі марки Сезон Зима',
'passage: Набор офисной бумаги цветной Magnat Color А4 80г/м.кв 4цветовХ50листов NEON',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
## Evaluation
### Metrics
#### Information Retrieval
* Datasets: `bm-full`, `core-uk-title`, `core-ru-title`, `core-uk-options`, `core-ru-options`, `options-uk-title`, `options-ru-title`, `options-uk-options`, `options-ru-options`, `rusisms-uk-title`, `rusisms-ru-title`, `rusisms-uk-options`, `rusisms-ru-options`, `rusisms_corrected-uk-title`, `rusisms_corrected-ru-title`, `rusisms_corrected-uk-options`, `rusisms_corrected-ru-options`, `core_typos-uk-title`, `core_typos-ru-title`, `core_typos-uk-options` and `core_typos-ru-options`
* Evaluated with [InformationRetrievalEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | bm-full | core-uk-title | core-ru-title | core-uk-options | core-ru-options | options-uk-title | options-ru-title | options-uk-options | options-ru-options | rusisms-uk-title | rusisms-ru-title | rusisms-uk-options | rusisms-ru-options | rusisms_corrected-uk-title | rusisms_corrected-ru-title | rusisms_corrected-uk-options | rusisms_corrected-ru-options | core_typos-uk-title | core_typos-ru-title | core_typos-uk-options | core_typos-ru-options |
|:-----------------|:-----------|:--------------|:--------------|:----------------|:----------------|:-----------------|:-----------------|:-------------------|:-------------------|:-----------------|:-----------------|:-------------------|:-------------------|:---------------------------|:---------------------------|:-----------------------------|:-----------------------------|:--------------------|:--------------------|:----------------------|:----------------------|
| dot_accuracy@1 | 0.4689 | 0.6314 | 0.6283 | 0.4928 | 0.4969 | 0.7795 | 0.7795 | 0.637 | 0.637 | 0.6413 | 0.6444 | 0.5048 | 0.5079 | 0.7152 | 0.7025 | 0.5665 | 0.5759 | 0.5431 | 0.5544 | 0.4004 | 0.4045 |
| dot_accuracy@3 | 0.6299 | 0.8501 | 0.8563 | 0.7454 | 0.77 | 0.9243 | 0.9287 | 0.8486 | 0.8441 | 0.7905 | 0.7714 | 0.6381 | 0.6508 | 0.8259 | 0.8165 | 0.7025 | 0.7089 | 0.7741 | 0.7721 | 0.6366 | 0.6458 |
| dot_accuracy@5 | 0.7222 | 0.9189 | 0.924 | 0.845 | 0.847 | 0.9621 | 0.971 | 0.9109 | 0.8931 | 0.8413 | 0.8222 | 0.6952 | 0.7206 | 0.8892 | 0.8734 | 0.769 | 0.7911 | 0.8491 | 0.846 | 0.7361 | 0.7433 |
| dot_accuracy@10 | 0.8153 | 0.9723 | 0.9733 | 0.9353 | 0.9312 | 0.9889 | 0.9955 | 0.9599 | 0.9532 | 0.8889 | 0.8825 | 0.8032 | 0.7937 | 0.9177 | 0.9146 | 0.8544 | 0.8734 | 0.9035 | 0.9086 | 0.846 | 0.848 |
| dot_precision@1 | 0.4689 | 0.6314 | 0.6283 | 0.4928 | 0.4969 | 0.7795 | 0.7795 | 0.637 | 0.637 | 0.6413 | 0.6444 | 0.5048 | 0.5079 | 0.7152 | 0.7025 | 0.5665 | 0.5759 | 0.5431 | 0.5544 | 0.4004 | 0.4045 |
| dot_precision@3 | 0.4743 | 0.6389 | 0.6468 | 0.5068 | 0.511 | 0.755 | 0.7572 | 0.6303 | 0.6362 | 0.6508 | 0.6529 | 0.5164 | 0.5249 | 0.7078 | 0.7068 | 0.5876 | 0.5781 | 0.538 | 0.5459 | 0.4144 | 0.4213 |
| dot_precision@5 | 0.484 | 0.6199 | 0.6304 | 0.5072 | 0.5068 | 0.7167 | 0.722 | 0.616 | 0.6138 | 0.6406 | 0.6444 | 0.5162 | 0.5219 | 0.7095 | 0.7 | 0.5905 | 0.5886 | 0.5216 | 0.5333 | 0.4107 | 0.4207 |
| dot_precision@10 | 0.4893 | 0.5194 | 0.5267 | 0.4586 | 0.4619 | 0.5592 | 0.5559 | 0.5042 | 0.4989 | 0.6124 | 0.6175 | 0.5108 | 0.521 | 0.6699 | 0.6715 | 0.5769 | 0.5854 | 0.4427 | 0.4483 | 0.3733 | 0.3737 |
| dot_recall@1 | 0.0111 | 0.0642 | 0.0642 | 0.0494 | 0.0494 | 0.1194 | 0.1193 | 0.0933 | 0.0944 | 0.0409 | 0.0435 | 0.0319 | 0.0322 | 0.0499 | 0.0482 | 0.0366 | 0.0372 | 0.0556 | 0.0567 | 0.0397 | 0.0406 |
| dot_recall@3 | 0.0337 | 0.1947 | 0.1969 | 0.1526 | 0.1534 | 0.3455 | 0.3471 | 0.279 | 0.2841 | 0.1203 | 0.1172 | 0.0828 | 0.0846 | 0.1281 | 0.1269 | 0.0987 | 0.0955 | 0.164 | 0.1672 | 0.1253 | 0.1269 |
| dot_recall@5 | 0.0575 | 0.3131 | 0.3188 | 0.2537 | 0.2531 | 0.5403 | 0.5424 | 0.4531 | 0.4498 | 0.1767 | 0.1731 | 0.1279 | 0.1289 | 0.1981 | 0.1961 | 0.1566 | 0.1517 | 0.2638 | 0.2706 | 0.206 | 0.2109 |
| dot_recall@10 | 0.1144 | 0.5139 | 0.5205 | 0.4524 | 0.4551 | 0.7987 | 0.7947 | 0.7071 | 0.701 | 0.3015 | 0.2976 | 0.2426 | 0.2461 | 0.3339 | 0.3376 | 0.2921 | 0.2976 | 0.4394 | 0.4452 | 0.3702 | 0.3709 |
| **dot_ndcg@10** | **0.4854** | **0.5824** | **0.5899** | **0.4953** | **0.4987** | **0.773** | **0.7721** | **0.6689** | **0.6653** | **0.6531** | **0.6565** | **0.5348** | **0.5427** | **0.7184** | **0.717** | **0.6076** | **0.6138** | **0.4961** | **0.5034** | **0.4044** | **0.4067** |
| dot_mrr@10 | 0.5668 | 0.7522 | 0.7535 | 0.6406 | 0.6457 | 0.8563 | 0.8602 | 0.7514 | 0.7476 | 0.7262 | 0.7233 | 0.5903 | 0.5954 | 0.7861 | 0.7751 | 0.6545 | 0.6651 | 0.6681 | 0.6771 | 0.5403 | 0.5483 |
| dot_map@100 | 0.3632 | 0.5633 | 0.5706 | 0.4997 | 0.4994 | 0.7474 | 0.7486 | 0.6486 | 0.6485 | 0.5739 | 0.582 | 0.4915 | 0.4952 | 0.6363 | 0.6389 | 0.5618 | 0.5643 | 0.4702 | 0.4772 | 0.4013 | 0.4023 |
#### Information Retrieval
* Datasets: `bm-full--matryoshka_dim-768--`, `bm-full--matryoshka_dim-512--`, `bm-full--matryoshka_dim-256--` and `bm-full--matryoshka_dim-128--`
* Evaluated with [InformationRetrievalEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | bm-full--matryoshka_dim-768-- | bm-full--matryoshka_dim-512-- | bm-full--matryoshka_dim-256-- | bm-full--matryoshka_dim-128-- |
|:----------------|:------------------------------|:------------------------------|:------------------------------|:------------------------------|
| dot_accuracy@1 | 0.4689 | 0.4692 | 0.4731 | 0.466 |
| dot_precision@1 | 0.4689 | 0.4692 | 0.4731 | 0.466 |
| dot_recall@1 | 0.0111 | 0.011 | 0.0112 | 0.011 |
| **dot_ndcg@1** | **0.4689** | **0.4692** | **0.4731** | **0.466** |
| dot_mrr@1 | 0.4689 | 0.4692 | 0.4731 | 0.466 |
| dot_map@100 | 0.3632 | 0.3613 | 0.3536 | 0.3346 |
## Training Details
### Training Dataset
#### rozetka_positive_pairs
* Dataset: rozetka_positive_pairs
* Size: 17,247,827 training samples
* Columns: query and text
* Approximate statistics based on the first 1000 samples:
| | query | text |
|:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
| type | string | string |
| details | - min: 6 tokens
- mean: 12.09 tokens
- max: 30 tokens
| - min: 8 tokens
- mean: 54.87 tokens
- max: 433 tokens
|
* Samples:
| query | text |
|:-----------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| query: xsiomi 9c скло | passage: Защитные стекла Назначение Для мобильных телефонов Цвет Черный Теги Теги Наличие рамки C рамкой Форм-фактор Плоское Клеевой слой По всей поверхности |
| query: xsiomi 9c скло | passage: Захисне скло Glass Full Glue для Xiaomi Redmi 9A/9C/10A (Чорний) |
| query: xsiomi 9c скло | passage: Захисне скло Призначення Для мобільних телефонів Колір Чорний Теги Теги Наявність рамки З рамкою Форм-фактор Плоске Клейовий шар По всій поверхні |
* Loss: sentence_transformers_training.model.matryoshka2d_loss.RZTKMatryoshka2dLoss with these parameters:
```json
{
"loss": "RZTKMultipleNegativesRankingLoss",
"n_layers_per_step": 1,
"last_layer_weight": 1.0,
"prior_layers_weight": 1.0,
"kl_div_weight": 1.0,
"kl_temperature": 0.3,
"matryoshka_dims": [
768,
512,
256,
128
],
"matryoshka_weights": [
1,
1,
1,
1
],
"n_dims_per_step": 1
}
```
### Evaluation Dataset
#### rozetka_positive_pairs
* Dataset: rozetka_positive_pairs
* Size: 198,271 evaluation samples
* Columns: query and text
* Approximate statistics based on the first 1000 samples:
| | query | text |
|:--------|:---------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
| type | string | string |
| details | - min: 6 tokens
- mean: 8.58 tokens
- max: 17 tokens
| - min: 8 tokens
- mean: 49.94 tokens
- max: 347 tokens
|
* Samples:
| query | text |
|:--------------------------------------------|:------------------------------------------------------------------------|
| query: создаем нейронную сеть | passage: Створюємо нейронну мережу |
| query: создаем нейронную сеть | passage: Научная и техническая литература Переплет Мягкий |
| query: создаем нейронную сеть | passage: Создаем нейронную сеть (1666498) |
* Loss: sentence_transformers_training.model.matryoshka2d_loss.RZTKMatryoshka2dLoss with these parameters:
```json
{
"loss": "RZTKMultipleNegativesRankingLoss",
"n_layers_per_step": 1,
"last_layer_weight": 1.0,
"prior_layers_weight": 1.0,
"kl_div_weight": 1.0,
"kl_temperature": 0.3,
"matryoshka_dims": [
768,
512,
256,
128
],
"matryoshka_weights": [
1,
1,
1,
1
],
"n_dims_per_step": 1
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: steps
- `per_device_train_batch_size`: 88
- `per_device_eval_batch_size`: 88
- `learning_rate`: 2e-05
- `num_train_epochs`: 7.0
- `warmup_ratio`: 0.1
- `bf16`: True
- `bf16_full_eval`: True
- `tf32`: True
- `dataloader_num_workers`: 8
- `load_best_model_at_end`: True
- `optim`: adafactor
- `push_to_hub`: True
- `hub_model_id`: yklymchuk-rztk/multilingual-e5-base-matryoshka2d-mnr-5
- `hub_private_repo`: True
- `prompts`: {'query': 'query: ', 'text': 'passage: '}
- `batch_sampler`: no_duplicates
#### All Hyperparameters
Click to expand
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: steps
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 88
- `per_device_eval_batch_size`: 88
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 2e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 7.0
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.1
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: True
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: True
- `fp16_full_eval`: False
- `tf32`: True
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: True
- `dataloader_num_workers`: 8
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: True
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adafactor
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: True
- `resume_from_checkpoint`: None
- `hub_model_id`: yklymchuk-rztk/multilingual-e5-base-matryoshka2d-mnr-5
- `hub_strategy`: every_save
- `hub_private_repo`: True
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `include_for_metrics`: []
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `average_tokens_across_devices`: False
- `prompts`: {'query': 'query: ', 'text': 'passage: '}
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional
- `ddp_static_graph`: False
- `ddp_comm_hook`: bf16
- `gradient_as_bucket_view`: False
- `num_proc`: 30
Click to expand
| Epoch | Step | Training Loss | Validation Loss | bm-full_dot_ndcg@10 | core-uk-title_dot_ndcg@10 | core-ru-title_dot_ndcg@10 | core-uk-options_dot_ndcg@10 | core-ru-options_dot_ndcg@10 | options-uk-title_dot_ndcg@10 | options-ru-title_dot_ndcg@10 | options-uk-options_dot_ndcg@10 | options-ru-options_dot_ndcg@10 | rusisms-uk-title_dot_ndcg@10 | rusisms-ru-title_dot_ndcg@10 | rusisms-uk-options_dot_ndcg@10 | rusisms-ru-options_dot_ndcg@10 | rusisms_corrected-uk-title_dot_ndcg@10 | rusisms_corrected-ru-title_dot_ndcg@10 | rusisms_corrected-uk-options_dot_ndcg@10 | rusisms_corrected-ru-options_dot_ndcg@10 | core_typos-uk-title_dot_ndcg@10 | core_typos-ru-title_dot_ndcg@10 | core_typos-uk-options_dot_ndcg@10 | core_typos-ru-options_dot_ndcg@10 | bm-full--matryoshka_dim-768--_dot_ndcg@1 | bm-full--matryoshka_dim-512--_dot_ndcg@1 | bm-full--matryoshka_dim-256--_dot_ndcg@1 | bm-full--matryoshka_dim-128--_dot_ndcg@1 |
|:----------:|:----------:|:-------------:|:---------------:|:-------------------:|:-------------------------:|:-------------------------:|:---------------------------:|:---------------------------:|:----------------------------:|:----------------------------:|:------------------------------:|:------------------------------:|:----------------------------:|:----------------------------:|:------------------------------:|:------------------------------:|:--------------------------------------:|:--------------------------------------:|:----------------------------------------:|:----------------------------------------:|:-------------------------------:|:-------------------------------:|:---------------------------------:|:---------------------------------:|:----------------------------------------:|:----------------------------------------:|:----------------------------------------:|:----------------------------------------:|
| 0.0350 | 1715 | 4.6833 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.0700 | 3430 | 4.4315 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.1050 | 5145 | 3.8514 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.1400 | 6860 | 2.9489 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.1750 | 8575 | 2.332 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.2100 | 10290 | 2.0101 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.2450 | 12005 | 1.8692 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.2800 | 13720 | 1.7073 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.3150 | 15435 | 1.6079 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.3500 | 17150 | 1.4682 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.3850 | 18865 | 1.4136 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.4200 | 20580 | 1.3448 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.4550 | 22295 | 1.3257 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.4900 | 24010 | 1.2269 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.5250 | 25725 | 1.1763 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.5600 | 27440 | 1.1717 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.5950 | 29155 | 1.0768 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.6300 | 30870 | 1.053 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.6650 | 32585 | 1.0062 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.7000 | 34300 | 0.9737 | 0.7655 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.7350 | 36015 | 0.914 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.7700 | 37730 | 0.9213 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.8050 | 39445 | 0.8538 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.8400 | 41160 | 0.8274 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.8750 | 42875 | 0.8191 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.9100 | 44590 | 0.8288 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.9450 | 46305 | 0.7787 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 0.9800 | 48020 | 0.7885 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.0151 | 49735 | 0.7577 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.0501 | 51450 | 0.7596 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.0851 | 53165 | 0.748 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.1201 | 54880 | 0.7131 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.1551 | 56595 | 0.709 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.1901 | 58310 | 0.729 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.2251 | 60025 | 0.6987 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.2601 | 61740 | 0.6788 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.2951 | 63455 | 0.6802 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.3301 | 65170 | 0.6565 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.3651 | 66885 | 0.6506 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.4001 | 68600 | 0.6539 | 0.4702 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.4351 | 70315 | 0.6499 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.4701 | 72030 | 0.6528 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.5051 | 73745 | 0.6416 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.5401 | 75460 | 0.6132 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.5751 | 77175 | 0.6404 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.6101 | 78890 | 0.6138 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.6451 | 80605 | 0.6155 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.6801 | 82320 | 0.5952 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.7151 | 84035 | 0.5892 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.7501 | 85750 | 0.5988 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.7851 | 87465 | 0.5972 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.8201 | 89180 | 0.5778 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.8551 | 90895 | 0.5546 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.8901 | 92610 | 0.5702 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.9251 | 94325 | 0.5651 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.9601 | 96040 | 0.5653 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 1.9951 | 97755 | 0.589 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.0301 | 99470 | 0.5746 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.0651 | 101185 | 0.5539 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.1001 | 102900 | 0.5652 | 0.4043 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.1351 | 104615 | 0.556 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.1701 | 106330 | 0.5547 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.2051 | 108045 | 0.5524 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.2401 | 109760 | 0.5568 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.2751 | 111475 | 0.5454 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.3101 | 113190 | 0.5612 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.3451 | 114905 | 0.5513 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.3801 | 116620 | 0.5475 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.4151 | 118335 | 0.5324 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.4501 | 120050 | 0.5408 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.4851 | 121765 | 0.5316 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.5201 | 123480 | 0.5473 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.5551 | 125195 | 0.5304 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.5901 | 126910 | 0.5392 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.6251 | 128625 | 0.5307 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.6601 | 130340 | 0.5427 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.6951 | 132055 | 0.5362 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.7301 | 133770 | 0.5301 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.7651 | 135485 | 0.5202 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.8001 | 137200 | 0.512 | 0.3902 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.8351 | 138915 | 0.5231 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.8701 | 140630 | 0.5093 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.9051 | 142345 | 0.5141 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.9401 | 144060 | 0.5199 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 2.9751 | 145775 | 0.5217 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.0102 | 147490 | 0.5325 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.0452 | 149205 | 0.5134 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.0802 | 150920 | 0.5251 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.1152 | 152635 | 0.5188 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.1502 | 154350 | 0.5299 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.1852 | 156065 | 0.5111 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.2202 | 157780 | 0.499 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.2552 | 159495 | 0.5044 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.2902 | 161210 | 0.5241 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.3252 | 162925 | 0.4969 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.3602 | 164640 | 0.5103 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.3952 | 166355 | 0.5093 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.4302 | 168070 | 0.5193 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.4652 | 169785 | 0.5075 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| **3.5002** | **171500** | **0.5096** | **0.3566** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** | **-** |
| 3.5352 | 173215 | 0.5009 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.5702 | 174930 | 0.5002 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.6052 | 176645 | 0.495 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.6402 | 178360 | 0.4962 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.6752 | 180075 | 0.4936 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.7102 | 181790 | 0.5068 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.7452 | 183505 | 0.4901 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.7802 | 185220 | 0.506 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.8152 | 186935 | 0.4885 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.8502 | 188650 | 0.5007 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.8852 | 190365 | 0.4907 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.9202 | 192080 | 0.5098 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.9552 | 193795 | 0.5088 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 3.9902 | 195510 | 0.4996 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.0252 | 197225 | 0.4998 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.0602 | 198940 | 0.4949 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.0952 | 200655 | 0.4928 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.1302 | 202370 | 0.5094 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.1652 | 204085 | 0.5117 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.2002 | 205800 | 0.5146 | 0.3682 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.2352 | 207515 | 0.4997 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.2703 | 209230 | 0.5029 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.3053 | 210945 | 0.5072 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.3403 | 212660 | 0.5031 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.3753 | 214375 | 0.4865 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.4103 | 216090 | 0.5038 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.4453 | 217805 | 0.4913 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.4803 | 219520 | 0.4805 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.5153 | 221235 | 0.4946 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.5503 | 222950 | 0.4918 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.5853 | 224665 | 0.4966 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.6203 | 226380 | 0.4993 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.6553 | 228095 | 0.4969 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.6903 | 229810 | 0.4803 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.7253 | 231525 | 0.47 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.7603 | 233240 | 0.4929 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.7953 | 234955 | 0.4971 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.8303 | 236670 | 0.4918 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.8653 | 238385 | 0.5 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.9003 | 240100 | 0.4998 | 0.3684 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.9353 | 241815 | 0.4893 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 4.9703 | 243530 | 0.5131 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.0053 | 245245 | 0.4984 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.0403 | 246960 | 0.4783 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.0753 | 248675 | 0.4942 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.1103 | 250390 | 0.5082 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.1453 | 252105 | 0.5115 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.1803 | 253820 | 0.4987 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.2153 | 255535 | 0.478 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.2503 | 257250 | 0.4914 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.2853 | 258965 | 0.4985 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.3203 | 260680 | 0.5017 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.3553 | 262395 | 0.4854 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.3903 | 264110 | 0.4868 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.4253 | 265825 | 0.4952 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.4603 | 267540 | 0.5083 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.4953 | 269255 | 0.4871 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.5303 | 270970 | 0.5057 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.5653 | 272685 | 0.4833 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.6003 | 274400 | 0.4879 | 0.3682 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.6353 | 276115 | 0.4952 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.6703 | 277830 | 0.5007 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.7053 | 279545 | 0.4818 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.7403 | 281260 | 0.4922 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.7753 | 282975 | 0.4861 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.8103 | 284690 | 0.4845 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.8453 | 286405 | 0.4846 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.8803 | 288120 | 0.4999 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.9153 | 289835 | 0.5012 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.9503 | 291550 | 0.496 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 5.9853 | 293265 | 0.508 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.0204 | 294980 | 0.501 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.0554 | 296695 | 0.4955 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.0904 | 298410 | 0.4924 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.1254 | 300125 | 0.509 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.1604 | 301840 | 0.5048 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.1954 | 303555 | 0.4965 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.2304 | 305270 | 0.5016 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.2654 | 306985 | 0.5104 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.3004 | 308700 | 0.5102 | 0.3593 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.3354 | 310415 | 0.4913 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.3704 | 312130 | 0.4941 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.4054 | 313845 | 0.5 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.4404 | 315560 | 0.5073 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.4754 | 317275 | 0.4937 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.5104 | 318990 | 0.4999 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.5454 | 320705 | 0.498 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.5804 | 322420 | 0.4886 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.6154 | 324135 | 0.4919 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.6504 | 325850 | 0.497 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.6854 | 327565 | 0.4949 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.7204 | 329280 | 0.494 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.7554 | 330995 | 0.4899 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.7904 | 332710 | 0.5003 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.8254 | 334425 | 0.502 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.8604 | 336140 | 0.4918 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.8954 | 337855 | 0.505 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.9304 | 339570 | 0.5136 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 6.9654 | 341285 | 0.5019 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| 7.0000 | 342979 | - | - | 0.4854 | 0.5824 | 0.5899 | 0.4953 | 0.4987 | 0.7730 | 0.7721 | 0.6689 | 0.6653 | 0.6531 | 0.6565 | 0.5348 | 0.5427 | 0.7184 | 0.7170 | 0.6076 | 0.6138 | 0.4961 | 0.5034 | 0.4044 | 0.4067 | 0.4689 | 0.4692 | 0.4731 | 0.4660 |
* The bold row denotes the saved checkpoint.