README.md

metadata

base_model: cointegrated/rubert-tiny2
datasets: []
language: []
library_name: sentence-transformers
metrics:
  - cosine_accuracy@1
  - cosine_accuracy@3
  - cosine_accuracy@5
  - cosine_precision@1
  - cosine_precision@3
  - cosine_precision@5
  - cosine_precision@10
  - cosine_recall@1
  - cosine_recall@3
  - cosine_recall@5
  - cosine_recall@10
  - cosine_ndcg@10
  - cosine_mrr@10
  - cosine_map@100
  - dot_accuracy@1
  - dot_accuracy@3
  - dot_accuracy@5
  - 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
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:1630
  - loss:MultipleNegativesRankingLoss
widget:
  - source_sentence: Сотруднику не войти в ЛК
    sentences:
      - >-
        При проблемах со входом в личный кабинет, прежде чем создавать заявку в
        поддержку, убедитесь, что заходите в ЛК на сайте https://company-x5.ru,
        указываете актуальные и верные логин и пароль. Если Вам неизвестен
        логин, обратитесь к руководителю (ДМ), он сможет посмотреть Ваш логин и
        сбросить пароль в веб-табеле. Для самостоятельного сброса пароля
        позвоните с вашего мобильного телефона на +7 (XXX) XXX XX XX, наберите
        добавочный номер 10100, нажмите * и подтвердите сброс пароля, нажав #.
        Обновленный пароль отправляется по SMS.
      - >-
        Рекомендуем уточнить наличие вакансий в регионе, куда планируется
        переезд. Обратиться к руководителю для формирования заявки в рамках
        процесса «Перевод» с заполнением атрибутов заявки.
      - >-
        Оформление отпуска без сохранения заработной платы возможно 2 способами:
        1. в разделе "Отпуска" в левом меню Личного кабинета сотрудника
        (https://company-x5.ru/vacations/plan?vp_page=1 ); 2. в разделе
        "Заявки", группа "Отпуск", плитка "Отпуск без сохранения ЗП".
        (https://company-x5.ru/requests/tiles/my/)
  - source_sentence: Как внести данные об автомобиле?
    sentences:
      - >-
        У Вас согласована Заявка на удаленную работу. График на удаленную работу
        ранее Вами не создавался. Вам необходимо зайти в личный кабинет, открыть
        вкладку "Удаленная работа" и создать себе график (от 1 до 6 месяцев).
        После создания, график перейдет к руководителю на согласование. А после
        согласования к Вам на подписание. После того, как процесс завершится,
        информирование прекратится.
      - >-
        Для внесения данных по личному автомобилю обратитесь, пожалуйста, к
        своему руководителю для создания заявки по теме "Изменение режима
        характера работы", подтема "Установка РХР и топливной карты". В
        комментариях опишите ситуацию и приложите ПТС, СТС, страховой полис и
        водительское удостоверение.
      - >-
        Выплата производится в случае утраты или повреждения жизненно
        необходимого недвижимого имущества работника, вследствие стихийных
        бедствий, пожара, кражи и т.д.;
  - source_sentence: не получается оформить внутреннего совместителя
    sentences:
      - Создайте, пожалуйста, обращение в ИТ поддержку на портале support
      - >-
        Вам необходимо обратиться в специалистам HR для того, чтобы они удалили
        мероприятие "работа на дому" и доп.соглашение, тогда статус текущей
        заявки станет «Не актуально» и появится возможность создать новую.
      - >-
        Оформление заявки на прием внутреннего совместителя производится
        руководителем магазина/отдела/РЦ, в который трудоустраивается сотрудник.
        При положительном решении он создает заявку по теме "Прием
        совместителей". Инструкция и шаблоны доступны по ссылке
        https://company-x5.ru/cms/zayavkaskillaz
  - source_sentence: Пропал календарь
    sentences:
      - >-
        При проблемах со входом в личный кабинет, прежде чем создавать заявку в
        поддержку, убедитесь, что заходите в ЛК на сайте https://company-x5.ru,
        указываете актуальные и верные логин и пароль. Если Вам неизвестен
        логин, обратитесь к руководителю (ДМ), он сможет посмотреть Ваш логин и
        сбросить пароль в веб-табеле. Для самостоятельного сброса пароля
        позвоните с вашего мобильного телефона на +7 (XXX) XXX XX XX, наберите
        добавочный номер 10100, нажмите * и подтвердите сброс пароля, нажав #.
        Обновленный пароль отправляется по SMS.
      - >-
        Доступ к программе "карьера" появляется спустя 4 месяца после
        трудоутстройства в компанию. Если по прошествии 4х месяцев раздел
        по-прежнему недоступен, обратитесь в поддержку
      - >-
        Вкладка "график работы" недоступна сотрудникам с организационным
        присвоением "Офис" и на данный момент доступна только сотрудникам
        розницы.
  - source_sentence: когда я получу деньги за отпуск
    sentences:
      - Создайте, пожалуйста, обращение в ИТ поддержку на портале support
      - >-
        Для изменения процента занятости сотруднику создайте, пожалуйста, заявку
        на сотрудника по теме "Изменение режима, характера работы"
      - >-
        Отпускные начисляются не позднее чем за три рабочих дня до даты начала
        отпуска.
model-index:
  - name: SentenceTransformer based on cointegrated/rubert-tiny2
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: test
          type: test
        metrics:
          - type: cosine_accuracy@1
            value: 0.7760736196319018
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.9337423312883436
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.9674846625766871
            name: Cosine Accuracy@5
          - type: cosine_precision@1
            value: 0.7760736196319018
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.3112474437627812
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.19349693251533742
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.09852760736196317
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.7760736196319018
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.9337423312883436
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.9674846625766871
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.9852760736196319
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.8898589364073973
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.85816851689551
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.8590198622778252
            name: Cosine Map@100
          - type: dot_accuracy@1
            value: 0.7760736196319018
            name: Dot Accuracy@1
          - type: dot_accuracy@3
            value: 0.9337423312883436
            name: Dot Accuracy@3
          - type: dot_accuracy@5
            value: 0.9674846625766871
            name: Dot Accuracy@5
          - type: dot_precision@1
            value: 0.7760736196319018
            name: Dot Precision@1
          - type: dot_precision@3
            value: 0.3112474437627812
            name: Dot Precision@3
          - type: dot_precision@5
            value: 0.19349693251533742
            name: Dot Precision@5
          - type: dot_precision@10
            value: 0.09852760736196317
            name: Dot Precision@10
          - type: dot_recall@1
            value: 0.7760736196319018
            name: Dot Recall@1
          - type: dot_recall@3
            value: 0.9337423312883436
            name: Dot Recall@3
          - type: dot_recall@5
            value: 0.9674846625766871
            name: Dot Recall@5
          - type: dot_recall@10
            value: 0.9852760736196319
            name: Dot Recall@10
          - type: dot_ndcg@10
            value: 0.8898589364073973
            name: Dot Ndcg@10
          - type: dot_mrr@10
            value: 0.85816851689551
            name: Dot Mrr@10
          - type: dot_map@100
            value: 0.8590198622778252
            name: Dot Map@100
          - type: cosine_accuracy@1
            value: 0.9895705521472392
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 1
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 1
            name: Cosine Accuracy@5
          - type: cosine_precision@1
            value: 0.9895705521472392
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.33333333333333326
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.2
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.9895705521472392
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 1
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 1
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.996150801110868
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.9947852760736197
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.9947852760736197
            name: Cosine Map@100
          - type: dot_accuracy@1
            value: 0.9895705521472392
            name: Dot Accuracy@1
          - type: dot_accuracy@3
            value: 1
            name: Dot Accuracy@3
          - type: dot_accuracy@5
            value: 1
            name: Dot Accuracy@5
          - type: dot_precision@1
            value: 0.9895705521472392
            name: Dot Precision@1
          - type: dot_precision@3
            value: 0.33333333333333326
            name: Dot Precision@3
          - type: dot_precision@5
            value: 0.2
            name: Dot Precision@5
          - type: dot_precision@10
            value: 0.1
            name: Dot Precision@10
          - type: dot_recall@1
            value: 0.9895705521472392
            name: Dot Recall@1
          - type: dot_recall@3
            value: 1
            name: Dot Recall@3
          - type: dot_recall@5
            value: 1
            name: Dot Recall@5
          - type: dot_recall@10
            value: 1
            name: Dot Recall@10
          - type: dot_ndcg@10
            value: 0.996150801110868
            name: Dot Ndcg@10
          - type: dot_mrr@10
            value: 0.9947852760736197
            name: Dot Mrr@10
          - type: dot_map@100
            value: 0.9947852760736197
            name: Dot Map@100

SentenceTransformer based on cointegrated/rubert-tiny2

This is a sentence-transformers model finetuned from cointegrated/rubert-tiny2. It maps sentences & paragraphs to a 312-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: cointegrated/rubert-tiny2
• Maximum Sequence Length: 2048 tokens
• Output Dimensionality: 312 tokens
• Similarity Function: Cosine Similarity

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 2048, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 312, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, '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:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("WpythonW/RUbert-tiny_custom_test_2")
# Run inference
sentences = [
    'когда я получу деньги за отпуск',
    'Отпускные начисляются не позднее чем за три рабочих дня до даты начала отпуска.',
    'Создайте, пожалуйста, обращение в ИТ поддержку на портале support',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 312]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Information Retrieval

• Dataset: test
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.7761
cosine_accuracy@3	0.9337
cosine_accuracy@5	0.9675
cosine_precision@1	0.7761
cosine_precision@3	0.3112
cosine_precision@5	0.1935
cosine_precision@10	0.0985
cosine_recall@1	0.7761
cosine_recall@3	0.9337
cosine_recall@5	0.9675
cosine_recall@10	0.9853
cosine_ndcg@10	0.8899
cosine_mrr@10	0.8582
cosine_map@100	0.859
dot_accuracy@1	0.7761
dot_accuracy@3	0.9337
dot_accuracy@5	0.9675
dot_precision@1	0.7761
dot_precision@3	0.3112
dot_precision@5	0.1935
dot_precision@10	0.0985
dot_recall@1	0.7761
dot_recall@3	0.9337
dot_recall@5	0.9675
dot_recall@10	0.9853
dot_ndcg@10	0.8899
dot_mrr@10	0.8582
dot_map@100	0.859

Information Retrieval

• Dataset: test
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.9896
cosine_accuracy@3	1.0
cosine_accuracy@5	1.0
cosine_precision@1	0.9896
cosine_precision@3	0.3333
cosine_precision@5	0.2
cosine_precision@10	0.1
cosine_recall@1	0.9896
cosine_recall@3	1.0
cosine_recall@5	1.0
cosine_recall@10	1.0
cosine_ndcg@10	0.9962
cosine_mrr@10	0.9948
cosine_map@100	0.9948
dot_accuracy@1	0.9896
dot_accuracy@3	1.0
dot_accuracy@5	1.0
dot_precision@1	0.9896
dot_precision@3	0.3333
dot_precision@5	0.2
dot_precision@10	0.1
dot_recall@1	0.9896
dot_recall@3	1.0
dot_recall@5	1.0
dot_recall@10	1.0
dot_ndcg@10	0.9962
dot_mrr@10	0.9948
dot_map@100	0.9948

Training Details

Training Dataset

Unnamed Dataset

• Size: 1,630 training samples
• Columns: sentence_0 and sentence_1
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1
  type	string	string
  details	min: 3 tokens mean: 12.4 tokens max: 74 tokens	min: 7 tokens mean: 61.91 tokens max: 371 tokens

• Samples:

  sentence_0	sentence_1
  Не отображается вкладка премия в ЛК. В консультации написали, что личный кабинет не передан на обслуживание в сервисную функцию HR Поддержку X5	Создайте, пожалуйста, обращение в ИТ поддержку на портале support
  как пересмотреть зарплату?	По данному вопросу Вы можете обратиться в кадровую службу, создав заявку "Консультация по HR вопросам"
  поменять телефон сотруднику	Кнопка "изменить номер" телефона находится в личном разделе в ЛК. Если доступа к ЛК нет, для смены номера телефона, обратитесь в поддержку

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 512
• per_device_eval_batch_size: 512
• num_train_epochs: 1200
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 512
• per_device_eval_batch_size: 512
• 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: 5e-05
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1
• num_train_epochs: 1200
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.0
• 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: False
• fp16: False
• fp16_opt_level: O1
• half_precision_backend: auto
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• local_rank: 0
• ddp_backend: None
• tpu_num_cores: None
• tpu_metrics_debug: False
• debug: []
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_prefetch_factor: None
• past_index: -1
• disable_tqdm: False
• remove_unused_columns: True
• label_names: None
• load_best_model_at_end: False
• 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: adamw_torch
• 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: False
• resume_from_checkpoint: None
• hub_model_id: None
• hub_strategy: every_save
• hub_private_repo: False
• hub_always_push: False
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• 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
• eval_use_gather_object: False
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin

Training Logs

Click to expand

Epoch	Step	Training Loss	test_cosine_map@100
1.0	4	-	0.1922
2.0	8	-	0.1922
3.0	12	-	0.1925
4.0	16	-	0.1927
5.0	20	-	0.1929
6.0	24	-	0.1931
7.0	28	-	0.1934
8.0	32	-	0.1942
9.0	36	-	0.1951
10.0	40	-	0.1960
11.0	44	-	0.1977
12.0	48	-	0.1993
13.0	52	-	0.2011
14.0	56	-	0.2024
15.0	60	-	0.2042
16.0	64	-	0.2047
17.0	68	-	0.2064
18.0	72	-	0.2081
19.0	76	-	0.2106
20.0	80	-	0.2123
21.0	84	-	0.2132
22.0	88	-	0.2148
23.0	92	-	0.2173
24.0	96	-	0.2200
25.0	100	-	0.2216
26.0	104	-	0.2241
27.0	108	-	0.2262
28.0	112	-	0.2286
29.0	116	-	0.2317
30.0	120	-	0.2339
31.0	124	-	0.2353
32.0	128	-	0.2392
33.0	132	-	0.2421
34.0	136	-	0.2442
35.0	140	-	0.2469
36.0	144	-	0.2501
37.0	148	-	0.2543
38.0	152	-	0.2565
39.0	156	-	0.2603
40.0	160	-	0.2643
41.0	164	-	0.2668
42.0	168	-	0.2688
43.0	172	-	0.2711
44.0	176	-	0.2743
45.0	180	-	0.2767
46.0	184	-	0.2810
47.0	188	-	0.2838
48.0	192	-	0.2869
49.0	196	-	0.2896
50.0	200	-	0.2940
51.0	204	-	0.2972
52.0	208	-	0.3012
53.0	212	-	0.3053
54.0	216	-	0.3072
55.0	220	-	0.3097
56.0	224	-	0.3133
57.0	228	-	0.3171
58.0	232	-	0.3220
59.0	236	-	0.3249
60.0	240	-	0.3274
61.0	244	-	0.3304
62.0	248	-	0.3336
63.0	252	-	0.3357
64.0	256	-	0.3398
65.0	260	-	0.3438
66.0	264	-	0.3463
67.0	268	-	0.3498
68.0	272	-	0.3535
69.0	276	-	0.3580
70.0	280	-	0.3606
71.0	284	-	0.3634
72.0	288	-	0.3654
73.0	292	-	0.3679
74.0	296	-	0.3723
75.0	300	-	0.3750
76.0	304	-	0.3781
77.0	308	-	0.3810
78.0	312	-	0.3840
79.0	316	-	0.3871
80.0	320	-	0.3914
81.0	324	-	0.3958
82.0	328	-	0.3991
83.0	332	-	0.4025
84.0	336	-	0.4053
85.0	340	-	0.4091
86.0	344	-	0.4125
87.0	348	-	0.4148
88.0	352	-	0.4176
89.0	356	-	0.4212
90.0	360	-	0.4240
91.0	364	-	0.4277
92.0	368	-	0.4315
93.0	372	-	0.4338
94.0	376	-	0.4363
95.0	380	-	0.4392
96.0	384	-	0.4423
97.0	388	-	0.4460
98.0	392	-	0.4487
99.0	396	-	0.4526
100.0	400	-	0.4563
101.0	404	-	0.4597
102.0	408	-	0.4644
103.0	412	-	0.4678
104.0	416	-	0.4707
105.0	420	-	0.4750
106.0	424	-	0.4791
107.0	428	-	0.4820
108.0	432	-	0.4847
109.0	436	-	0.4895
110.0	440	-	0.4916
111.0	444	-	0.4961
112.0	448	-	0.4990
113.0	452	-	0.5032
114.0	456	-	0.5065
115.0	460	-	0.5093
116.0	464	-	0.5135
117.0	468	-	0.5175
118.0	472	-	0.5199
119.0	476	-	0.5243
120.0	480	-	0.5266
121.0	484	-	0.5297
122.0	488	-	0.5324
123.0	492	-	0.5353
124.0	496	-	0.5378
125.0	500	4.6316	0.5404
126.0	504	-	0.5449
127.0	508	-	0.5473
128.0	512	-	0.5500
129.0	516	-	0.5535
130.0	520	-	0.5553
131.0	524	-	0.5570
132.0	528	-	0.5597
133.0	532	-	0.5625
134.0	536	-	0.5660
135.0	540	-	0.5693
136.0	544	-	0.5713
137.0	548	-	0.5745
138.0	552	-	0.5767
139.0	556	-	0.5798
140.0	560	-	0.5835
141.0	564	-	0.5853
142.0	568	-	0.5863
143.0	572	-	0.5909
144.0	576	-	0.5933
145.0	580	-	0.5970
146.0	584	-	0.5995
147.0	588	-	0.6008
148.0	592	-	0.6040
149.0	596	-	0.6073
150.0	600	-	0.6097
151.0	604	-	0.6121
152.0	608	-	0.6163
153.0	612	-	0.6178
154.0	616	-	0.6205
155.0	620	-	0.6223
156.0	624	-	0.6242
157.0	628	-	0.6265
158.0	632	-	0.6290
159.0	636	-	0.6326
160.0	640	-	0.6347
161.0	644	-	0.6370
162.0	648	-	0.6400
163.0	652	-	0.6422
164.0	656	-	0.6436
165.0	660	-	0.6460
166.0	664	-	0.6473
167.0	668	-	0.6497
168.0	672	-	0.6515
169.0	676	-	0.6545
170.0	680	-	0.6574
171.0	684	-	0.6595
172.0	688	-	0.6616
173.0	692	-	0.6639
174.0	696	-	0.6658
175.0	700	-	0.6676
176.0	704	-	0.6697
177.0	708	-	0.6713
178.0	712	-	0.6746
179.0	716	-	0.6765
180.0	720	-	0.6784
181.0	724	-	0.6806
182.0	728	-	0.6820
183.0	732	-	0.6838
184.0	736	-	0.6867
185.0	740	-	0.6882
186.0	744	-	0.6913
187.0	748	-	0.6930
188.0	752	-	0.6943
189.0	756	-	0.6979
190.0	760	-	0.6982
191.0	764	-	0.7008
192.0	768	-	0.7038
193.0	772	-	0.7059
194.0	776	-	0.7062
195.0	780	-	0.7083
196.0	784	-	0.7112
197.0	788	-	0.7139
198.0	792	-	0.7163
199.0	796	-	0.7181
200.0	800	-	0.7188
201.0	804	-	0.7208
202.0	808	-	0.7223
203.0	812	-	0.7245
204.0	816	-	0.7264
205.0	820	-	0.7296
206.0	824	-	0.7325
207.0	828	-	0.7340
208.0	832	-	0.7362
209.0	836	-	0.7373
210.0	840	-	0.7394
211.0	844	-	0.7416
212.0	848	-	0.7420
213.0	852	-	0.7434
214.0	856	-	0.7444
215.0	860	-	0.7466
216.0	864	-	0.7479
217.0	868	-	0.7523
218.0	872	-	0.7540
219.0	876	-	0.7553
220.0	880	-	0.7558
221.0	884	-	0.7586
222.0	888	-	0.7596
223.0	892	-	0.7613
224.0	896	-	0.7637
225.0	900	-	0.7652
226.0	904	-	0.7667
227.0	908	-	0.7682
228.0	912	-	0.7698
229.0	916	-	0.7715
230.0	920	-	0.7729
231.0	924	-	0.7757
232.0	928	-	0.7767
233.0	932	-	0.7783
234.0	936	-	0.7802
235.0	940	-	0.7814
236.0	944	-	0.7835
237.0	948	-	0.7859
238.0	952	-	0.7874
239.0	956	-	0.7887
240.0	960	-	0.7903
241.0	964	-	0.7927
242.0	968	-	0.7940
243.0	972	-	0.7958
244.0	976	-	0.7973
245.0	980	-	0.7991
246.0	984	-	0.8009
247.0	988	-	0.8021
248.0	992	-	0.8035
249.0	996	-	0.8043
250.0	1000	3.2323	0.8057
251.0	1004	-	0.8071
252.0	1008	-	0.8088
253.0	1012	-	0.8104
254.0	1016	-	0.8112
255.0	1020	-	0.8123
256.0	1024	-	0.8135
257.0	1028	-	0.8154
258.0	1032	-	0.8167
259.0	1036	-	0.8174
260.0	1040	-	0.8187
261.0	1044	-	0.8187
262.0	1048	-	0.8210
263.0	1052	-	0.8216
264.0	1056	-	0.8242
265.0	1060	-	0.8260
266.0	1064	-	0.8267
267.0	1068	-	0.8278
268.0	1072	-	0.8294
269.0	1076	-	0.8309
270.0	1080	-	0.8319
271.0	1084	-	0.8325
272.0	1088	-	0.8346
273.0	1092	-	0.8353
274.0	1096	-	0.8362
275.0	1100	-	0.8373
276.0	1104	-	0.8385
277.0	1108	-	0.8392
278.0	1112	-	0.8405
279.0	1116	-	0.8431
280.0	1120	-	0.8453
281.0	1124	-	0.8464
282.0	1128	-	0.8480
283.0	1132	-	0.8476
284.0	1136	-	0.8491
285.0	1140	-	0.8509
286.0	1144	-	0.8508
287.0	1148	-	0.8513
288.0	1152	-	0.8525
289.0	1156	-	0.8534
290.0	1160	-	0.8543
291.0	1164	-	0.8554
292.0	1168	-	0.8572
293.0	1172	-	0.8590
1.0	4	-	0.8591
2.0	8	-	0.8591
3.0	12	-	0.8591
4.0	16	-	0.8591
5.0	20	-	0.8591
6.0	24	-	0.8591
7.0	28	-	0.8592
8.0	32	-	0.8592
9.0	36	-	0.8589
10.0	40	-	0.8593
11.0	44	-	0.8587
12.0	48	-	0.8590
13.0	52	-	0.8591
14.0	56	-	0.8591
15.0	60	-	0.8593
16.0	64	-	0.8593
17.0	68	-	0.8595
18.0	72	-	0.8598
19.0	76	-	0.8602
20.0	80	-	0.8606
21.0	84	-	0.8614
22.0	88	-	0.8617
23.0	92	-	0.8617
24.0	96	-	0.8621
25.0	100	-	0.8622
26.0	104	-	0.8626
27.0	108	-	0.8626
28.0	112	-	0.8626
29.0	116	-	0.8629
30.0	120	-	0.8629
31.0	124	-	0.8629
32.0	128	-	0.8629
33.0	132	-	0.8628
34.0	136	-	0.8625
35.0	140	-	0.8626
36.0	144	-	0.8628
37.0	148	-	0.8628
38.0	152	-	0.8630
39.0	156	-	0.8635
40.0	160	-	0.8635
41.0	164	-	0.8642
42.0	168	-	0.8646
43.0	172	-	0.8649
44.0	176	-	0.8654
45.0	180	-	0.8658
46.0	184	-	0.8662
47.0	188	-	0.8666
48.0	192	-	0.8676
49.0	196	-	0.8676
50.0	200	-	0.8677
51.0	204	-	0.8681
52.0	208	-	0.8680
53.0	212	-	0.8677
54.0	216	-	0.8682
55.0	220	-	0.8683
56.0	224	-	0.8687
57.0	228	-	0.8687
58.0	232	-	0.8687
59.0	236	-	0.8689
60.0	240	-	0.8690
61.0	244	-	0.8697
62.0	248	-	0.8700
63.0	252	-	0.8706
64.0	256	-	0.8706
65.0	260	-	0.8709
66.0	264	-	0.8711
67.0	268	-	0.8711
68.0	272	-	0.8716
69.0	276	-	0.8717
70.0	280	-	0.8728
71.0	284	-	0.8728
72.0	288	-	0.8729
73.0	292	-	0.8732
74.0	296	-	0.8734
75.0	300	-	0.8741
76.0	304	-	0.8736
77.0	308	-	0.8739
78.0	312	-	0.8742
79.0	316	-	0.8743
80.0	320	-	0.8744
81.0	324	-	0.8749
82.0	328	-	0.8750
83.0	332	-	0.8760
84.0	336	-	0.8758
85.0	340	-	0.8765
86.0	344	-	0.8771
87.0	348	-	0.8771
88.0	352	-	0.8768
89.0	356	-	0.8778
90.0	360	-	0.8778
91.0	364	-	0.8784
92.0	368	-	0.8793
93.0	372	-	0.8795
94.0	376	-	0.8796
95.0	380	-	0.8801
96.0	384	-	0.8803
97.0	388	-	0.8806
98.0	392	-	0.8811
99.0	396	-	0.8815
100.0	400	-	0.8814
101.0	404	-	0.8822
102.0	408	-	0.8829
103.0	412	-	0.8827
104.0	416	-	0.8827
105.0	420	-	0.8838
106.0	424	-	0.8836
107.0	428	-	0.8841
108.0	432	-	0.8848
109.0	436	-	0.8853
110.0	440	-	0.8857
111.0	444	-	0.8858
112.0	448	-	0.8863
113.0	452	-	0.8868
114.0	456	-	0.8877
115.0	460	-	0.8887
116.0	464	-	0.8887
117.0	468	-	0.8883
118.0	472	-	0.8883
119.0	476	-	0.8887
120.0	480	-	0.8888
121.0	484	-	0.8902
122.0	488	-	0.8900
123.0	492	-	0.8906
124.0	496	-	0.8908
125.0	500	2.6383	0.8909
126.0	504	-	0.8913
127.0	508	-	0.8917
128.0	512	-	0.8922
129.0	516	-	0.8923
130.0	520	-	0.8924
131.0	524	-	0.8924
132.0	528	-	0.8927
133.0	532	-	0.8925
134.0	536	-	0.8931
135.0	540	-	0.8941
136.0	544	-	0.8953
137.0	548	-	0.8957
138.0	552	-	0.8966
139.0	556	-	0.8980
140.0	560	-	0.8985
141.0	564	-	0.8979
142.0	568	-	0.8993
143.0	572	-	0.8987
144.0	576	-	0.8998
145.0	580	-	0.8994
146.0	584	-	0.9002
147.0	588	-	0.9006
148.0	592	-	0.9016
149.0	596	-	0.9025
150.0	600	-	0.9026
151.0	604	-	0.9030
152.0	608	-	0.9035
153.0	612	-	0.9043
154.0	616	-	0.9050
155.0	620	-	0.9060
156.0	624	-	0.9060
157.0	628	-	0.9067
158.0	632	-	0.9062
159.0	636	-	0.9065
160.0	640	-	0.9083
161.0	644	-	0.9085
162.0	648	-	0.9090
163.0	652	-	0.9093
164.0	656	-	0.9098
165.0	660	-	0.9101
166.0	664	-	0.9104
167.0	668	-	0.9111
168.0	672	-	0.9121
169.0	676	-	0.9123
170.0	680	-	0.9130
171.0	684	-	0.9134
172.0	688	-	0.9135
173.0	692	-	0.9139
174.0	696	-	0.9145
175.0	700	-	0.9143
176.0	704	-	0.9146
177.0	708	-	0.9155
178.0	712	-	0.9164
179.0	716	-	0.9185
180.0	720	-	0.9192
181.0	724	-	0.9192
182.0	728	-	0.9192
183.0	732	-	0.9205
184.0	736	-	0.9208
185.0	740	-	0.9210
186.0	744	-	0.9216
187.0	748	-	0.9216
188.0	752	-	0.9219
189.0	756	-	0.9218
190.0	760	-	0.9221
191.0	764	-	0.9233
192.0	768	-	0.9240
193.0	772	-	0.9253
194.0	776	-	0.9255
195.0	780	-	0.9256
196.0	784	-	0.9259
197.0	788	-	0.9260
198.0	792	-	0.9268
199.0	796	-	0.9271
200.0	800	-	0.9271
201.0	804	-	0.9273
202.0	808	-	0.9282
203.0	812	-	0.9280
204.0	816	-	0.9278
205.0	820	-	0.9290
206.0	824	-	0.9294
207.0	828	-	0.9302
208.0	832	-	0.9307
209.0	836	-	0.9304
210.0	840	-	0.9304
211.0	844	-	0.9316
212.0	848	-	0.9320
213.0	852	-	0.9325
214.0	856	-	0.9332
215.0	860	-	0.9335
216.0	864	-	0.9348
217.0	868	-	0.9359
218.0	872	-	0.9362
219.0	876	-	0.9362
220.0	880	-	0.9362
221.0	884	-	0.9362
222.0	888	-	0.9363
223.0	892	-	0.9368
224.0	896	-	0.9374
225.0	900	-	0.9381
226.0	904	-	0.9378
227.0	908	-	0.9378
228.0	912	-	0.9376
229.0	916	-	0.9373
230.0	920	-	0.9380
231.0	924	-	0.9381
232.0	928	-	0.9386
233.0	932	-	0.9399
234.0	936	-	0.9404
235.0	940	-	0.9402
236.0	944	-	0.9406
237.0	948	-	0.9406
238.0	952	-	0.9404
239.0	956	-	0.9410
240.0	960	-	0.9412
241.0	964	-	0.9412
242.0	968	-	0.9423
243.0	972	-	0.9429
244.0	976	-	0.9429
245.0	980	-	0.9432
246.0	984	-	0.9432
247.0	988	-	0.9439
248.0	992	-	0.9449
249.0	996	-	0.9452
250.0	1000	2.507	0.9461
251.0	1004	-	0.9464
252.0	1008	-	0.9464
253.0	1012	-	0.9460
254.0	1016	-	0.9456
255.0	1020	-	0.9471
256.0	1024	-	0.9468
257.0	1028	-	0.9472
258.0	1032	-	0.9476
259.0	1036	-	0.9481
260.0	1040	-	0.9490
261.0	1044	-	0.9488
262.0	1048	-	0.9488
263.0	1052	-	0.9478
264.0	1056	-	0.9475
265.0	1060	-	0.9485
266.0	1064	-	0.9490
267.0	1068	-	0.9487
268.0	1072	-	0.9484
269.0	1076	-	0.9490
270.0	1080	-	0.9495
271.0	1084	-	0.9506
272.0	1088	-	0.9513
273.0	1092	-	0.9518
274.0	1096	-	0.9522
275.0	1100	-	0.9526
276.0	1104	-	0.9522
277.0	1108	-	0.9526
278.0	1112	-	0.9531
279.0	1116	-	0.9537
280.0	1120	-	0.9533
281.0	1124	-	0.9523
282.0	1128	-	0.9544
283.0	1132	-	0.9546
284.0	1136	-	0.9553
285.0	1140	-	0.9554
286.0	1144	-	0.9565
287.0	1148	-	0.9568
288.0	1152	-	0.9569
289.0	1156	-	0.9569
290.0	1160	-	0.9567
291.0	1164	-	0.9568
292.0	1168	-	0.9574
293.0	1172	-	0.9574
294.0	1176	-	0.9574
295.0	1180	-	0.9580
296.0	1184	-	0.9586
297.0	1188	-	0.9588
298.0	1192	-	0.9594
299.0	1196	-	0.9596
300.0	1200	-	0.9602
301.0	1204	-	0.9604
302.0	1208	-	0.9598
303.0	1212	-	0.9605
304.0	1216	-	0.9608
305.0	1220	-	0.9614
306.0	1224	-	0.9620
307.0	1228	-	0.9621
308.0	1232	-	0.9630
309.0	1236	-	0.9633
310.0	1240	-	0.9644
311.0	1244	-	0.9644
312.0	1248	-	0.9643
313.0	1252	-	0.9644
314.0	1256	-	0.9644
315.0	1260	-	0.9646
316.0	1264	-	0.9661
317.0	1268	-	0.9665
318.0	1272	-	0.9664
319.0	1276	-	0.9666
320.0	1280	-	0.9673
321.0	1284	-	0.9681
322.0	1288	-	0.9681
323.0	1292	-	0.9684
324.0	1296	-	0.9685
325.0	1300	-	0.9686
326.0	1304	-	0.9681
327.0	1308	-	0.9686
328.0	1312	-	0.9684
329.0	1316	-	0.9685
330.0	1320	-	0.9688
331.0	1324	-	0.9690
332.0	1328	-	0.9691
333.0	1332	-	0.9695
334.0	1336	-	0.9701
335.0	1340	-	0.9714
336.0	1344	-	0.9713
337.0	1348	-	0.9719
338.0	1352	-	0.9720
339.0	1356	-	0.9720
340.0	1360	-	0.9720
341.0	1364	-	0.9720
342.0	1368	-	0.9725
343.0	1372	-	0.9728
344.0	1376	-	0.9725
345.0	1380	-	0.9723
346.0	1384	-	0.9727
347.0	1388	-	0.9723
348.0	1392	-	0.9729
349.0	1396	-	0.9735
350.0	1400	-	0.9735
351.0	1404	-	0.9732
352.0	1408	-	0.9739
353.0	1412	-	0.9742
354.0	1416	-	0.9747
355.0	1420	-	0.9744
356.0	1424	-	0.9744
357.0	1428	-	0.9744
358.0	1432	-	0.9743
359.0	1436	-	0.9740
360.0	1440	-	0.9742
361.0	1444	-	0.9739
362.0	1448	-	0.9736
363.0	1452	-	0.9746
364.0	1456	-	0.9752
365.0	1460	-	0.9752
366.0	1464	-	0.9752
367.0	1468	-	0.9748
368.0	1472	-	0.9748
369.0	1476	-	0.9749
370.0	1480	-	0.9755
371.0	1484	-	0.9753
372.0	1488	-	0.9759
373.0	1492	-	0.9760
374.0	1496	-	0.9755
375.0	1500	2.391	0.9755
376.0	1504	-	0.9757
377.0	1508	-	0.9757
378.0	1512	-	0.9760
379.0	1516	-	0.9762
380.0	1520	-	0.9760
381.0	1524	-	0.9762
382.0	1528	-	0.9761
383.0	1532	-	0.9761
384.0	1536	-	0.9770
385.0	1540	-	0.9774
386.0	1544	-	0.9777
387.0	1548	-	0.9780
388.0	1552	-	0.9774
389.0	1556	-	0.9768
390.0	1560	-	0.9780
391.0	1564	-	0.9789
392.0	1568	-	0.9789
393.0	1572	-	0.9786
394.0	1576	-	0.9786
395.0	1580	-	0.9783
396.0	1584	-	0.9789
397.0	1588	-	0.9790
398.0	1592	-	0.9787
399.0	1596	-	0.9788
400.0	1600	-	0.9782
401.0	1604	-	0.9782
402.0	1608	-	0.9782
403.0	1612	-	0.9782
404.0	1616	-	0.9788
405.0	1620	-	0.9789
406.0	1624	-	0.9789
407.0	1628	-	0.9793
408.0	1632	-	0.9794
409.0	1636	-	0.9797
410.0	1640	-	0.9803
411.0	1644	-	0.9800
412.0	1648	-	0.9796
413.0	1652	-	0.9799
414.0	1656	-	0.9799
415.0	1660	-	0.9796
416.0	1664	-	0.9797
417.0	1668	-	0.9797
418.0	1672	-	0.9800
419.0	1676	-	0.9803
420.0	1680	-	0.9809
421.0	1684	-	0.9806
422.0	1688	-	0.9809
423.0	1692	-	0.9812
424.0	1696	-	0.9810
425.0	1700	-	0.9806
426.0	1704	-	0.9806
427.0	1708	-	0.9799
428.0	1712	-	0.9796
429.0	1716	-	0.9802
430.0	1720	-	0.9802
431.0	1724	-	0.9810
432.0	1728	-	0.9810
433.0	1732	-	0.9807
434.0	1736	-	0.9810
435.0	1740	-	0.9813
436.0	1744	-	0.9816
437.0	1748	-	0.9820
438.0	1752	-	0.9816
439.0	1756	-	0.9816
440.0	1760	-	0.9813
441.0	1764	-	0.9820
442.0	1768	-	0.9823
443.0	1772	-	0.9820
444.0	1776	-	0.9823
445.0	1780	-	0.9826
446.0	1784	-	0.9823
447.0	1788	-	0.9832
448.0	1792	-	0.9832
449.0	1796	-	0.9832
450.0	1800	-	0.9835
451.0	1804	-	0.9835
452.0	1808	-	0.9835
453.0	1812	-	0.9835
454.0	1816	-	0.9835
455.0	1820	-	0.9835
456.0	1824	-	0.9838
457.0	1828	-	0.9838
458.0	1832	-	0.9841
459.0	1836	-	0.9841
460.0	1840	-	0.9841
461.0	1844	-	0.9841
462.0	1848	-	0.9844
463.0	1852	-	0.9850
464.0	1856	-	0.9844
465.0	1860	-	0.9841
466.0	1864	-	0.9844
467.0	1868	-	0.9850
468.0	1872	-	0.9853
469.0	1876	-	0.9850
470.0	1880	-	0.9856
471.0	1884	-	0.9856
472.0	1888	-	0.9856
473.0	1892	-	0.9853
474.0	1896	-	0.9856
475.0	1900	-	0.9850
476.0	1904	-	0.9850
477.0	1908	-	0.9850
478.0	1912	-	0.9850
479.0	1916	-	0.9853
480.0	1920	-	0.9856
481.0	1924	-	0.9859
482.0	1928	-	0.9862
483.0	1932	-	0.9862
484.0	1936	-	0.9862
485.0	1940	-	0.9862
486.0	1944	-	0.9859
487.0	1948	-	0.9859
488.0	1952	-	0.9856
489.0	1956	-	0.9859
490.0	1960	-	0.9859
491.0	1964	-	0.9859
492.0	1968	-	0.9856
493.0	1972	-	0.9856
494.0	1976	-	0.9856
495.0	1980	-	0.9856
496.0	1984	-	0.9862
497.0	1988	-	0.9862
498.0	1992	-	0.9856
499.0	1996	-	0.9856
500.0	2000	2.3269	0.9856
501.0	2004	-	0.9856
502.0	2008	-	0.9856
503.0	2012	-	0.9859
504.0	2016	-	0.9862
505.0	2020	-	0.9866
506.0	2024	-	0.9866
507.0	2028	-	0.9869
508.0	2032	-	0.9869
509.0	2036	-	0.9869
510.0	2040	-	0.9875
511.0	2044	-	0.9875
512.0	2048	-	0.9875
513.0	2052	-	0.9872
514.0	2056	-	0.9872
515.0	2060	-	0.9869
516.0	2064	-	0.9869
517.0	2068	-	0.9866
518.0	2072	-	0.9866
519.0	2076	-	0.9862
520.0	2080	-	0.9866
521.0	2084	-	0.9866
522.0	2088	-	0.9862
523.0	2092	-	0.9866
524.0	2096	-	0.9862
525.0	2100	-	0.9866
526.0	2104	-	0.9872
527.0	2108	-	0.9878
528.0	2112	-	0.9878
529.0	2116	-	0.9881
530.0	2120	-	0.9881
531.0	2124	-	0.9881
532.0	2128	-	0.9878
533.0	2132	-	0.9878
534.0	2136	-	0.9878
535.0	2140	-	0.9878
536.0	2144	-	0.9875
537.0	2148	-	0.9878
538.0	2152	-	0.9872
539.0	2156	-	0.9869
540.0	2160	-	0.9872
541.0	2164	-	0.9875
542.0	2168	-	0.9878
543.0	2172	-	0.9878
544.0	2176	-	0.9881
545.0	2180	-	0.9888
546.0	2184	-	0.9894
547.0	2188	-	0.9894
548.0	2192	-	0.9897
549.0	2196	-	0.9897
550.0	2200	-	0.9897
551.0	2204	-	0.9897
552.0	2208	-	0.9897
553.0	2212	-	0.9894
554.0	2216	-	0.9891
555.0	2220	-	0.9888
556.0	2224	-	0.9884
557.0	2228	-	0.9884
558.0	2232	-	0.9884
559.0	2236	-	0.9888
560.0	2240	-	0.9891
561.0	2244	-	0.9891
562.0	2248	-	0.9894
563.0	2252	-	0.9897
564.0	2256	-	0.9897
565.0	2260	-	0.9897
566.0	2264	-	0.9900
567.0	2268	-	0.9903
568.0	2272	-	0.9900
569.0	2276	-	0.9903
570.0	2280	-	0.9900
571.0	2284	-	0.9900
572.0	2288	-	0.9900
573.0	2292	-	0.9900
574.0	2296	-	0.9903
575.0	2300	-	0.9903
576.0	2304	-	0.9903
577.0	2308	-	0.9903
578.0	2312	-	0.9903
579.0	2316	-	0.9897
580.0	2320	-	0.9897
581.0	2324	-	0.9897
582.0	2328	-	0.9897
583.0	2332	-	0.9900
584.0	2336	-	0.9900
585.0	2340	-	0.9900
586.0	2344	-	0.9904
587.0	2348	-	0.9904
588.0	2352	-	0.9904
589.0	2356	-	0.9901
590.0	2360	-	0.9901
591.0	2364	-	0.9898
592.0	2368	-	0.9898
593.0	2372	-	0.9898
594.0	2376	-	0.9901
595.0	2380	-	0.9901
596.0	2384	-	0.9901
597.0	2388	-	0.9901
598.0	2392	-	0.9901
599.0	2396	-	0.9904
600.0	2400	-	0.9904
601.0	2404	-	0.9904
602.0	2408	-	0.9904
603.0	2412	-	0.9904
604.0	2416	-	0.9907
605.0	2420	-	0.9904
606.0	2424	-	0.9904
607.0	2428	-	0.9904
608.0	2432	-	0.9904
609.0	2436	-	0.9904
610.0	2440	-	0.9904
611.0	2444	-	0.9907
612.0	2448	-	0.9907
613.0	2452	-	0.9907
614.0	2456	-	0.9907
615.0	2460	-	0.9907
616.0	2464	-	0.9907
617.0	2468	-	0.9910
618.0	2472	-	0.9910
619.0	2476	-	0.9910
620.0	2480	-	0.9910
621.0	2484	-	0.9913
622.0	2488	-	0.9910
623.0	2492	-	0.9907
624.0	2496	-	0.9907
625.0	2500	2.2939	0.9907
626.0	2504	-	0.9907
627.0	2508	-	0.9907
628.0	2512	-	0.9907
629.0	2516	-	0.9910
630.0	2520	-	0.9910
631.0	2524	-	0.9910
632.0	2528	-	0.9910
633.0	2532	-	0.9910
634.0	2536	-	0.9913
635.0	2540	-	0.9916
636.0	2544	-	0.9916
637.0	2548	-	0.9913
638.0	2552	-	0.9910
639.0	2556	-	0.9910
640.0	2560	-	0.9910
641.0	2564	-	0.9910
642.0	2568	-	0.9910
643.0	2572	-	0.9913
644.0	2576	-	0.9916
645.0	2580	-	0.9916
646.0	2584	-	0.9916
647.0	2588	-	0.9916
648.0	2592	-	0.9916
649.0	2596	-	0.9919
650.0	2600	-	0.9919
651.0	2604	-	0.9916
652.0	2608	-	0.9916
653.0	2612	-	0.9919
654.0	2616	-	0.9919
655.0	2620	-	0.9919
656.0	2624	-	0.9916
657.0	2628	-	0.9916
658.0	2632	-	0.9916
659.0	2636	-	0.9916
660.0	2640	-	0.9919
661.0	2644	-	0.9922
662.0	2648	-	0.9922
663.0	2652	-	0.9922
664.0	2656	-	0.9922
665.0	2660	-	0.9919
666.0	2664	-	0.9922
667.0	2668	-	0.9922
668.0	2672	-	0.9925
669.0	2676	-	0.9928
670.0	2680	-	0.9925
671.0	2684	-	0.9928
672.0	2688	-	0.9925
673.0	2692	-	0.9925
674.0	2696	-	0.9928
675.0	2700	-	0.9928
676.0	2704	-	0.9931
677.0	2708	-	0.9931
678.0	2712	-	0.9931
679.0	2716	-	0.9928
680.0	2720	-	0.9925
681.0	2724	-	0.9922
682.0	2728	-	0.9922
683.0	2732	-	0.9922
684.0	2736	-	0.9922
685.0	2740	-	0.9922
686.0	2744	-	0.9922
687.0	2748	-	0.9925
688.0	2752	-	0.9931
689.0	2756	-	0.9931
690.0	2760	-	0.9935
691.0	2764	-	0.9935
692.0	2768	-	0.9935
693.0	2772	-	0.9931
694.0	2776	-	0.9931
695.0	2780	-	0.9931
696.0	2784	-	0.9928
697.0	2788	-	0.9928
698.0	2792	-	0.9925
699.0	2796	-	0.9925
700.0	2800	-	0.9928
701.0	2804	-	0.9931
702.0	2808	-	0.9931
703.0	2812	-	0.9931
704.0	2816	-	0.9931
705.0	2820	-	0.9931
706.0	2824	-	0.9928
707.0	2828	-	0.9931
708.0	2832	-	0.9928
709.0	2836	-	0.9928
710.0	2840	-	0.9928
711.0	2844	-	0.9925
712.0	2848	-	0.9922
713.0	2852	-	0.9922
714.0	2856	-	0.9922
715.0	2860	-	0.9922
716.0	2864	-	0.9922
717.0	2868	-	0.9922
718.0	2872	-	0.9928
719.0	2876	-	0.9928
720.0	2880	-	0.9928
721.0	2884	-	0.9928
722.0	2888	-	0.9931
723.0	2892	-	0.9928
724.0	2896	-	0.9928
725.0	2900	-	0.9928
726.0	2904	-	0.9931
727.0	2908	-	0.9931
728.0	2912	-	0.9931
729.0	2916	-	0.9931
730.0	2920	-	0.9928
731.0	2924	-	0.9928
732.0	2928	-	0.9931
733.0	2932	-	0.9931
734.0	2936	-	0.9928
735.0	2940	-	0.9928
736.0	2944	-	0.9931
737.0	2948	-	0.9931
738.0	2952	-	0.9928
739.0	2956	-	0.9928
740.0	2960	-	0.9928
741.0	2964	-	0.9928
742.0	2968	-	0.9928
743.0	2972	-	0.9928
744.0	2976	-	0.9935
745.0	2980	-	0.9935
746.0	2984	-	0.9935
747.0	2988	-	0.9935
748.0	2992	-	0.9935
749.0	2996	-	0.9935
750.0	3000	2.2749	0.9935
751.0	3004	-	0.9935
752.0	3008	-	0.9938
753.0	3012	-	0.9938
754.0	3016	-	0.9938
755.0	3020	-	0.9941
756.0	3024	-	0.9938
757.0	3028	-	0.9938
758.0	3032	-	0.9938
759.0	3036	-	0.9938
760.0	3040	-	0.9938
761.0	3044	-	0.9938
762.0	3048	-	0.9938
763.0	3052	-	0.9939
764.0	3056	-	0.9942
765.0	3060	-	0.9942
766.0	3064	-	0.9939
767.0	3068	-	0.9939
768.0	3072	-	0.9942
769.0	3076	-	0.9939
770.0	3080	-	0.9939
771.0	3084	-	0.9938
772.0	3088	-	0.9938
773.0	3092	-	0.9938
774.0	3096	-	0.9938
775.0	3100	-	0.9938
776.0	3104	-	0.9938
777.0	3108	-	0.9935
778.0	3112	-	0.9935
779.0	3116	-	0.9935
780.0	3120	-	0.9938
781.0	3124	-	0.9938
782.0	3128	-	0.9935
783.0	3132	-	0.9935
784.0	3136	-	0.9935
785.0	3140	-	0.9931
786.0	3144	-	0.9931
787.0	3148	-	0.9931
788.0	3152	-	0.9931
789.0	3156	-	0.9931
790.0	3160	-	0.9931
791.0	3164	-	0.9931
792.0	3168	-	0.9935
793.0	3172	-	0.9935
794.0	3176	-	0.9935
795.0	3180	-	0.9935
796.0	3184	-	0.9935
797.0	3188	-	0.9933
798.0	3192	-	0.9933
799.0	3196	-	0.9936
800.0	3200	-	0.9936
801.0	3204	-	0.9933
802.0	3208	-	0.9935
803.0	3212	-	0.9938
804.0	3216	-	0.9935
805.0	3220	-	0.9931
806.0	3224	-	0.9936
807.0	3228	-	0.9936
808.0	3232	-	0.9939
809.0	3236	-	0.9942
810.0	3240	-	0.9945
811.0	3244	-	0.9945
812.0	3248	-	0.9945
813.0	3252	-	0.9945
814.0	3256	-	0.9942
815.0	3260	-	0.9939
816.0	3264	-	0.9942
817.0	3268	-	0.9939
818.0	3272	-	0.9942
819.0	3276	-	0.9942
820.0	3280	-	0.9942
821.0	3284	-	0.9945
822.0	3288	-	0.9945
823.0	3292	-	0.9945
824.0	3296	-	0.9945
825.0	3300	-	0.9945
826.0	3304	-	0.9945
827.0	3308	-	0.9945
828.0	3312	-	0.9945
829.0	3316	-	0.9945
830.0	3320	-	0.9945
831.0	3324	-	0.9945
832.0	3328	-	0.9945
833.0	3332	-	0.9945
834.0	3336	-	0.9948
835.0	3340	-	0.9948
836.0	3344	-	0.9948
837.0	3348	-	0.9948
838.0	3352	-	0.9948
839.0	3356	-	0.9948
840.0	3360	-	0.9948
841.0	3364	-	0.9948
842.0	3368	-	0.9948
843.0	3372	-	0.9948
844.0	3376	-	0.9945
845.0	3380	-	0.9945
846.0	3384	-	0.9948
847.0	3388	-	0.9948
848.0	3392	-	0.9948
849.0	3396	-	0.9948
850.0	3400	-	0.9948
851.0	3404	-	0.9948
852.0	3408	-	0.9948
853.0	3412	-	0.9948
854.0	3416	-	0.9948
855.0	3420	-	0.9948
856.0	3424	-	0.9948
857.0	3428	-	0.9945
858.0	3432	-	0.9945
859.0	3436	-	0.9945
860.0	3440	-	0.9945
861.0	3444	-	0.9945
862.0	3448	-	0.9948
863.0	3452	-	0.9948
864.0	3456	-	0.9948
865.0	3460	-	0.9948
866.0	3464	-	0.9948
867.0	3468	-	0.9948
868.0	3472	-	0.9948
869.0	3476	-	0.9948
870.0	3480	-	0.9948
871.0	3484	-	0.9948
872.0	3488	-	0.9948
873.0	3492	-	0.9948
874.0	3496	-	0.9948
875.0	3500	2.268	0.9948
876.0	3504	-	0.9948
877.0	3508	-	0.9948
878.0	3512	-	0.9948
879.0	3516	-	0.9948
880.0	3520	-	0.9948
881.0	3524	-	0.9948
882.0	3528	-	0.9948
883.0	3532	-	0.9948
884.0	3536	-	0.9948
885.0	3540	-	0.9948
886.0	3544	-	0.9948
887.0	3548	-	0.9948
888.0	3552	-	0.9948
889.0	3556	-	0.9948
890.0	3560	-	0.9948
891.0	3564	-	0.9948
892.0	3568	-	0.9948
893.0	3572	-	0.9948
894.0	3576	-	0.9948
895.0	3580	-	0.9948
896.0	3584	-	0.9948
897.0	3588	-	0.9948
898.0	3592	-	0.9948
899.0	3596	-	0.9948
900.0	3600	-	0.9948
901.0	3604	-	0.9948
902.0	3608	-	0.9948
903.0	3612	-	0.9948
904.0	3616	-	0.9948
905.0	3620	-	0.9948
906.0	3624	-	0.9948
907.0	3628	-	0.9948
908.0	3632	-	0.9948
909.0	3636	-	0.9948
910.0	3640	-	0.9948
911.0	3644	-	0.9948
912.0	3648	-	0.9948
913.0	3652	-	0.9948
914.0	3656	-	0.9948
915.0	3660	-	0.9948
916.0	3664	-	0.9948
917.0	3668	-	0.9948
918.0	3672	-	0.9948
919.0	3676	-	0.9948
920.0	3680	-	0.9948
921.0	3684	-	0.9948
922.0	3688	-	0.9948
923.0	3692	-	0.9948
924.0	3696	-	0.9948
925.0	3700	-	0.9948
926.0	3704	-	0.9948
927.0	3708	-	0.9948
928.0	3712	-	0.9948
929.0	3716	-	0.9948
930.0	3720	-	0.9948
931.0	3724	-	0.9948
932.0	3728	-	0.9948
933.0	3732	-	0.9948
934.0	3736	-	0.9948
935.0	3740	-	0.9948
936.0	3744	-	0.9948
937.0	3748	-	0.9948
938.0	3752	-	0.9948
939.0	3756	-	0.9948
940.0	3760	-	0.9948
941.0	3764	-	0.9948
942.0	3768	-	0.9948
943.0	3772	-	0.9948
944.0	3776	-	0.9948
945.0	3780	-	0.9948
946.0	3784	-	0.9948
947.0	3788	-	0.9948
948.0	3792	-	0.9948
949.0	3796	-	0.9948
950.0	3800	-	0.9948
951.0	3804	-	0.9948
952.0	3808	-	0.9948
953.0	3812	-	0.9948
954.0	3816	-	0.9948
955.0	3820	-	0.9948
956.0	3824	-	0.9948
957.0	3828	-	0.9948
958.0	3832	-	0.9948
959.0	3836	-	0.9948
960.0	3840	-	0.9948
961.0	3844	-	0.9948
962.0	3848	-	0.9948
963.0	3852	-	0.9948
964.0	3856	-	0.9948
965.0	3860	-	0.9948
966.0	3864	-	0.9948
967.0	3868	-	0.9948
968.0	3872	-	0.9948
969.0	3876	-	0.9948
970.0	3880	-	0.9948
971.0	3884	-	0.9948
972.0	3888	-	0.9948
973.0	3892	-	0.9948
974.0	3896	-	0.9948
975.0	3900	-	0.9948
976.0	3904	-	0.9948
977.0	3908	-	0.9948
978.0	3912	-	0.9948
979.0	3916	-	0.9948
980.0	3920	-	0.9948
981.0	3924	-	0.9948
982.0	3928	-	0.9948
983.0	3932	-	0.9948
984.0	3936	-	0.9948
985.0	3940	-	0.9948
986.0	3944	-	0.9948
987.0	3948	-	0.9948
988.0	3952	-	0.9948
989.0	3956	-	0.9948
990.0	3960	-	0.9948
991.0	3964	-	0.9948
992.0	3968	-	0.9948
993.0	3972	-	0.9948
994.0	3976	-	0.9948
995.0	3980	-	0.9948
996.0	3984	-	0.9948
997.0	3988	-	0.9948
998.0	3992	-	0.9948
999.0	3996	-	0.9948
1000.0	4000	2.265	0.9948
1001.0	4004	-	0.9948
1002.0	4008	-	0.9948
1003.0	4012	-	0.9948
1004.0	4016	-	0.9948
1005.0	4020	-	0.9948
1006.0	4024	-	0.9948
1007.0	4028	-	0.9948
1008.0	4032	-	0.9948
1009.0	4036	-	0.9948
1010.0	4040	-	0.9948
1011.0	4044	-	0.9948
1012.0	4048	-	0.9948
1013.0	4052	-	0.9948
1014.0	4056	-	0.9948
1015.0	4060	-	0.9948
1016.0	4064	-	0.9948
1017.0	4068	-	0.9948
1018.0	4072	-	0.9948
1019.0	4076	-	0.9948
1020.0	4080	-	0.9948
1021.0	4084	-	0.9948
1022.0	4088	-	0.9948
1023.0	4092	-	0.9948
1024.0	4096	-	0.9948
1025.0	4100	-	0.9948
1026.0	4104	-	0.9948
1027.0	4108	-	0.9948
1028.0	4112	-	0.9948
1029.0	4116	-	0.9948
1030.0	4120	-	0.9948
1031.0	4124	-	0.9948
1032.0	4128	-	0.9948
1033.0	4132	-	0.9948
1034.0	4136	-	0.9948
1035.0	4140	-	0.9948
1036.0	4144	-	0.9948
1037.0	4148	-	0.9948
1038.0	4152	-	0.9948
1039.0	4156	-	0.9948
1040.0	4160	-	0.9948
1041.0	4164	-	0.9948
1042.0	4168	-	0.9948
1043.0	4172	-	0.9948
1044.0	4176	-	0.9948
1045.0	4180	-	0.9948
1046.0	4184	-	0.9948
1047.0	4188	-	0.9948
1048.0	4192	-	0.9948
1049.0	4196	-	0.9948
1050.0	4200	-	0.9948
1051.0	4204	-	0.9948
1052.0	4208	-	0.9948
1053.0	4212	-	0.9948
1054.0	4216	-	0.9948
1055.0	4220	-	0.9948
1056.0	4224	-	0.9948
1057.0	4228	-	0.9948
1058.0	4232	-	0.9948
1059.0	4236	-	0.9948
1060.0	4240	-	0.9948
1061.0	4244	-	0.9948
1062.0	4248	-	0.9948
1063.0	4252	-	0.9948
1064.0	4256	-	0.9948
1065.0	4260	-	0.9948
1066.0	4264	-	0.9948
1067.0	4268	-	0.9948
1068.0	4272	-	0.9948
1069.0	4276	-	0.9948
1070.0	4280	-	0.9948
1071.0	4284	-	0.9948
1072.0	4288	-	0.9948
1073.0	4292	-	0.9948
1074.0	4296	-	0.9948
1075.0	4300	-	0.9948
1076.0	4304	-	0.9948
1077.0	4308	-	0.9948
1078.0	4312	-	0.9948
1079.0	4316	-	0.9948
1080.0	4320	-	0.9948
1081.0	4324	-	0.9948
1082.0	4328	-	0.9948
1083.0	4332	-	0.9948
1084.0	4336	-	0.9948
1085.0	4340	-	0.9948
1086.0	4344	-	0.9948
1087.0	4348	-	0.9948
1088.0	4352	-	0.9948
1089.0	4356	-	0.9948
1090.0	4360	-	0.9948
1091.0	4364	-	0.9948
1092.0	4368	-	0.9948
1093.0	4372	-	0.9948
1094.0	4376	-	0.9948
1095.0	4380	-	0.9948
1096.0	4384	-	0.9948
1097.0	4388	-	0.9948
1098.0	4392	-	0.9948
1099.0	4396	-	0.9948
1100.0	4400	-	0.9948
1101.0	4404	-	0.9948
1102.0	4408	-	0.9948
1103.0	4412	-	0.9948
1104.0	4416	-	0.9948
1105.0	4420	-	0.9948
1106.0	4424	-	0.9948
1107.0	4428	-	0.9948
1108.0	4432	-	0.9948
1109.0	4436	-	0.9948
1110.0	4440	-	0.9948
1111.0	4444	-	0.9948
1112.0	4448	-	0.9948
1113.0	4452	-	0.9948
1114.0	4456	-	0.9948
1115.0	4460	-	0.9948
1116.0	4464	-	0.9948
1117.0	4468	-	0.9948
1118.0	4472	-	0.9948
1119.0	4476	-	0.9948
1120.0	4480	-	0.9948
1121.0	4484	-	0.9948
1122.0	4488	-	0.9948
1123.0	4492	-	0.9948
1124.0	4496	-	0.9948
1125.0	4500	2.2654	0.9948
1126.0	4504	-	0.9948
1127.0	4508	-	0.9948
1128.0	4512	-	0.9948
1129.0	4516	-	0.9948
1130.0	4520	-	0.9948
1131.0	4524	-	0.9948
1132.0	4528	-	0.9948
1133.0	4532	-	0.9948
1134.0	4536	-	0.9948
1135.0	4540	-	0.9948
1136.0	4544	-	0.9948
1137.0	4548	-	0.9948
1138.0	4552	-	0.9948
1139.0	4556	-	0.9948
1140.0	4560	-	0.9948
1141.0	4564	-	0.9948
1142.0	4568	-	0.9948
1143.0	4572	-	0.9948
1144.0	4576	-	0.9948
1145.0	4580	-	0.9948
1146.0	4584	-	0.9948
1147.0	4588	-	0.9948
1148.0	4592	-	0.9948
1149.0	4596	-	0.9948
1150.0	4600	-	0.9948
1151.0	4604	-	0.9948
1152.0	4608	-	0.9948
1153.0	4612	-	0.9948
1154.0	4616	-	0.9948
1155.0	4620	-	0.9948
1156.0	4624	-	0.9948
1157.0	4628	-	0.9948
1158.0	4632	-	0.9948
1159.0	4636	-	0.9948
1160.0	4640	-	0.9948
1161.0	4644	-	0.9948
1162.0	4648	-	0.9948
1163.0	4652	-	0.9948
1164.0	4656	-	0.9948
1165.0	4660	-	0.9948
1166.0	4664	-	0.9948
1167.0	4668	-	0.9948
1168.0	4672	-	0.9948
1169.0	4676	-	0.9948
1170.0	4680	-	0.9948
1171.0	4684	-	0.9948
1172.0	4688	-	0.9948
1173.0	4692	-	0.9948
1174.0	4696	-	0.9948
1175.0	4700	-	0.9948
1176.0	4704	-	0.9948
1177.0	4708	-	0.9948
1178.0	4712	-	0.9948
1179.0	4716	-	0.9948
1180.0	4720	-	0.9948
1181.0	4724	-	0.9948
1182.0	4728	-	0.9948
1183.0	4732	-	0.9948
1184.0	4736	-	0.9948
1185.0	4740	-	0.9948
1186.0	4744	-	0.9948
1187.0	4748	-	0.9948
1188.0	4752	-	0.9948
1189.0	4756	-	0.9948
1190.0	4760	-	0.9948
1191.0	4764	-	0.9948
1192.0	4768	-	0.9948
1193.0	4772	-	0.9948
1194.0	4776	-	0.9948
1195.0	4780	-	0.9948
1196.0	4784	-	0.9948
1197.0	4788	-	0.9948
1198.0	4792	-	0.9948
1199.0	4796	-	0.9948
1200.0	4800	-	0.9948

Framework Versions

• Python: 3.10.14
• Sentence Transformers: 3.0.1
• Transformers: 4.44.0
• PyTorch: 2.4.0
• Accelerate: 0.34.2
• Datasets: 2.21.0
• Tokenizers: 0.19.1

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply}, 
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}