sergeyzh commited on
Commit
3fc5dc6
1 Parent(s): 192b8be

Upload 9 files

Browse files
1_Pooling/config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 768,
3
+ "pooling_mode_cls_token": true,
4
+ "pooling_mode_mean_tokens": false,
5
+ "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false
7
+ }
README.md CHANGED
@@ -1,3 +1,93 @@
1
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  license: mit
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ language:
3
+ - ru
4
+
5
+ pipeline_tag: sentence-similarity
6
+
7
+ tags:
8
+ - russian
9
+ - pretraining
10
+ - embeddings
11
+ - feature-extraction
12
+ - sentence-similarity
13
+ - sentence-transformers
14
+ - transformers
15
+
16
  license: mit
17
+
18
  ---
19
+
20
+ ## Базовый Bert для Semantic text similarity (STS) на GPU
21
+
22
+ Качественная модель BERT для расчетов эмбедингов предложений на русском языке. Модель основана на [cointegrated/LaBSE-en-ru](https://huggingface.co/cointegrated/LaBSE-en-ru) - имеет аналогичный размеры контекста (512), ембединга (768) и быстродействие. Является второй и лучшей по качеству моделью в серии BERT-sts.
23
+
24
+ На STS и близких задачах (PI, NLI, SA, TI) для русского языка конкурирует по качеству с моделью [intfloat/multilingual-e5-large](https://huggingface.co/intfloat/multilingual-e5-large) (но быстрее в 1.8 раза и меньше на 77%). Оптимальна для использования в составе RAG LLMs при инференсе на GPU.
25
+
26
+ ## Использование модели с библиотекой `transformers`:
27
+
28
+ ```python
29
+ # pip install transformers sentencepiece
30
+ import torch
31
+ from transformers import AutoTokenizer, AutoModel
32
+ tokenizer = AutoTokenizer.from_pretrained("sergeyzh/LaBSE-ru-sts")
33
+ model = AutoModel.from_pretrained("sergeyzh/LaBSE-ru-sts")
34
+ # model.cuda() # uncomment it if you have a GPU
35
+
36
+ def embed_bert_cls(text, model, tokenizer):
37
+ t = tokenizer(text, padding=True, truncation=True, return_tensors='pt')
38
+ with torch.no_grad():
39
+ model_output = model(**{k: v.to(model.device) for k, v in t.items()})
40
+ embeddings = model_output.last_hidden_state[:, 0, :]
41
+ embeddings = torch.nn.functional.normalize(embeddings)
42
+ return embeddings[0].cpu().numpy()
43
+
44
+ print(embed_bert_cls('привет мир', model, tokenizer).shape)
45
+ # (768,)
46
+ ```
47
+
48
+ ## Использование с `sentence_transformers`:
49
+ ```Python
50
+ from sentence_transformers import SentenceTransformer, util
51
+
52
+ model = SentenceTransformer('sergeyzh/LaBSE-ru-sts')
53
+
54
+ sentences = ["привет мир", "hello world", "здравствуй вселенная"]
55
+ embeddings = model.encode(sentences)
56
+ print(util.dot_score(embeddings, embeddings))
57
+ ```
58
+
59
+ ## Метрики
60
+ Оценки модели на бенчмарке [encodechka](https://github.com/avidale/encodechka):
61
+
62
+ | Модель | STS | PI | NLI | SA | TI |
63
+ |:---------------------------------|:---------:|:---------:|:---------:|:---------:|:---------:|
64
+ | [intfloat/multilingual-e5-large](https://huggingface.co/intfloat/multilingual-e5-large) | 0.862 | 0.727 | 0.473 | 0.810 | 0.979 |
65
+ | **sergeyzh/LaBSE-ru-sts** | **0.845** | **0.737** | **0.481** | **0.805** | **0.957** |
66
+ | [sergeyzh/rubert-tiny-sts](https://huggingface.co/sergeyzh/rubert-tiny-sts) | 0.797 | 0.702 | 0.453 | 0.778 | 0.946 |
67
+ | [Tochka-AI/ruRoPEBert-e5-base-512](https://huggingface.co/Tochka-AI/ruRoPEBert-e5-base-512) | 0.793 | 0.704 | 0.457 | 0.803 | 0.970 |
68
+ | [cointegrated/LaBSE-en-ru](https://huggingface.co/cointegrated/LaBSE-en-ru) | 0.794 | 0.659 | 0.431 | 0.761 | 0.946 |
69
+ | [cointegrated/rubert-tiny2](https://huggingface.co/cointegrated/rubert-tiny2) | 0.750 | 0.651 | 0.417 | 0.737 | 0.937 |
70
+
71
+ **Задачи:**
72
+
73
+ - Semantic text similarity (**STS**);
74
+ - Paraphrase identification (**PI**);
75
+ - Natural language inference (**NLI**);
76
+ - Sentiment analysis (**SA**);
77
+ - Toxicity identification (**TI**).
78
+
79
+ ## Быстродействие и размеры
80
+
81
+ | Модель | CPU | GPU | size | dim | n_ctx | n_vocab |
82
+ |:---------------------------------|----------:|----------:|----------:|----------:|----------:|----------:|
83
+ | [intfloat/multilingual-e5-large](https://huggingface.co/intfloat/multilingual-e5-large) | 149.026 | 15.629 | 2136 | 1024 | 514 | 250002 |
84
+ | **sergeyzh/LaBSE-ru-sts** |**42.835** | **8.561** | **490** | **768** | **512** | **55083** |
85
+ | [sergeyzh/rubert-tiny-sts](https://huggingface.co/sergeyzh/rubert-tiny-sts) | 3.208 | 2.866 | 111 | 312 | 2048 | 83828 |
86
+ | [Tochka-AI/ruRoPEBert-e5-base-512](https://huggingface.co/Tochka-AI/ruRoPEBert-e5-base-512) | 43.314 | 9.338 | 530 | 768 | 512 | 69382 |
87
+ | [cointegrated/LaBSE-en-ru](https://huggingface.co/cointegrated/LaBSE-en-ru) | 42.867 | 8.549 | 490 | 768 | 512 | 55083 |
88
+ | [cointegrated/rubert-tiny2](https://huggingface.co/cointegrated/rubert-tiny2) | 3.212 | 2.850 | 111 | 312 | 2048 | 83828 |
89
+
90
+
91
+ ## Связанные ресурсы
92
+ Вопросы использования модели обсуждаются в [русскоязычном чате NLP](https://t.me/natural_language_processing).
93
+
config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "sergeyzh/LaBSE-ru-sts",
3
+ "architectures": [
4
+ "BertForPreTraining"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "directionality": "bidi",
9
+ "gradient_checkpointing": false,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 3072,
15
+ "layer_norm_eps": 1e-12,
16
+ "max_position_embeddings": 512,
17
+ "model_type": "bert",
18
+ "num_attention_heads": 12,
19
+ "num_hidden_layers": 12,
20
+ "pad_token_id": 0,
21
+ "pooler_fc_size": 768,
22
+ "pooler_num_attention_heads": 12,
23
+ "pooler_num_fc_layers": 3,
24
+ "pooler_size_per_head": 128,
25
+ "pooler_type": "first_token_transform",
26
+ "position_embedding_type": "absolute",
27
+ "torch_dtype": "float32",
28
+ "transformers_version": "4.38.1",
29
+ "type_vocab_size": 2,
30
+ "use_cache": true,
31
+ "vocab_size": 55083
32
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:897438c44abad8096eee6b1508547d2eca917bb640588efae88669cdc852ac46
3
+ size 515999444
modules.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "idx": 0,
4
+ "name": "0",
5
+ "path": "",
6
+ "type": "sentence_transformers.models.Transformer"
7
+ },
8
+ {
9
+ "idx": 1,
10
+ "name": "1",
11
+ "path": "1_Pooling",
12
+ "type": "sentence_transformers.models.Pooling"
13
+ },
14
+ {
15
+ "idx": 2,
16
+ "name": "2",
17
+ "path": "2_Normalize",
18
+ "type": "sentence_transformers.models.Normalize"
19
+ }
20
+ ]
sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 256,
3
+ "do_lower_case": false
4
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": false,
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "never_split": null,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "strip_accents": null,
54
+ "tokenize_chinese_chars": true,
55
+ "tokenizer_class": "BertTokenizer",
56
+ "unk_token": "[UNK]"
57
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff