nreimers commited on
Commit
1243633
1 Parent(s): bdb6aff

Add new SentenceTransformer model.

Browse files
1_Pooling/config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 768,
3
+ "pooling_mode_cls_token": false,
4
+ "pooling_mode_mean_tokens": true,
5
+ "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false
7
+ }
README.md ADDED
@@ -0,0 +1,133 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ pipeline_tag: sentence-similarity
3
+ tags:
4
+ - sentence-transformers
5
+ - feature-extraction
6
+ - sentence-similarity
7
+ - transformers
8
+ - transformers
9
+ - transformers
10
+ - transformers
11
+ - transformers
12
+ - transformers
13
+ - transformers
14
+ - transformers
15
+ - transformers
16
+ - transformers
17
+ - transformers
18
+ - transformers
19
+ - transformers
20
+ - transformers
21
+ - transformers
22
+ - transformers
23
+ - transformers
24
+ - transformers
25
+ - transformers
26
+ - transformers
27
+ - transformers
28
+ - transformers
29
+ - transformers
30
+ - transformers
31
+ - transformers
32
+ - transformers
33
+ - transformers
34
+ - transformers
35
+ ---
36
+
37
+ # sentence-transformers/distilbert-multilingual-nli-stsb-quora-ranking
38
+
39
+ This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.
40
+
41
+
42
+
43
+ ## Usage (Sentence-Transformers)
44
+
45
+ Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:
46
+
47
+ ```
48
+ pip install -U sentence-transformers
49
+ ```
50
+
51
+ Then you can use the model like this:
52
+
53
+ ```python
54
+ from sentence_transformers import SentenceTransformer
55
+ sentences = ["This is an example sentence", "Each sentence is converted"]
56
+
57
+ model = SentenceTransformer('sentence-transformers/distilbert-multilingual-nli-stsb-quora-ranking')
58
+ embeddings = model.encode(sentences)
59
+ print(embeddings)
60
+ ```
61
+
62
+
63
+
64
+ ## Usage (HuggingFace Transformers)
65
+ Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.
66
+
67
+ ```python
68
+ from transformers import AutoTokenizer, AutoModel
69
+ import torch
70
+
71
+
72
+ #Mean Pooling - Take attention mask into account for correct averaging
73
+ def mean_pooling(model_output, attention_mask):
74
+ token_embeddings = model_output[0] #First element of model_output contains all token embeddings
75
+ input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
76
+ return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
77
+
78
+
79
+ # Sentences we want sentence embeddings for
80
+ sentences = ['This is an example sentence', 'Each sentence is converted']
81
+
82
+ # Load model from HuggingFace Hub
83
+ tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/distilbert-multilingual-nli-stsb-quora-ranking')
84
+ model = AutoModel.from_pretrained('sentence-transformers/distilbert-multilingual-nli-stsb-quora-ranking')
85
+
86
+ # Tokenize sentences
87
+ encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
88
+
89
+ # Compute token embeddings
90
+ with torch.no_grad():
91
+ model_output = model(**encoded_input)
92
+
93
+ # Perform pooling. In this case, max pooling.
94
+ sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
95
+
96
+ print("Sentence embeddings:")
97
+ print(sentence_embeddings)
98
+ ```
99
+
100
+
101
+
102
+ ## Evaluation Results
103
+
104
+
105
+
106
+ For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name=sentence-transformers/distilbert-multilingual-nli-stsb-quora-ranking)
107
+
108
+
109
+
110
+ ## Full Model Architecture
111
+ ```
112
+ SentenceTransformer(
113
+ (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: DistilBertModel
114
+ (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})
115
+ )
116
+ ```
117
+
118
+ ## Citing & Authors
119
+
120
+ This model was trained by [sentence-transformers](https://www.sbert.net/).
121
+
122
+ If you find this model helpful, feel free to cite our publication [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://arxiv.org/abs/1908.10084):
123
+ ```bibtex
124
+ @inproceedings{reimers-2019-sentence-bert,
125
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
126
+ author = "Reimers, Nils and Gurevych, Iryna",
127
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
128
+ month = "11",
129
+ year = "2019",
130
+ publisher = "Association for Computational Linguistics",
131
+ url = "http://arxiv.org/abs/1908.10084",
132
+ }
133
+ ```
config.json CHANGED
@@ -1,4 +1,5 @@
1
  {
 
2
  "activation": "gelu",
3
  "architectures": [
4
  "DistilBertModel"
@@ -18,5 +19,6 @@
18
  "seq_classif_dropout": 0.2,
19
  "sinusoidal_pos_embds": false,
20
  "tie_weights_": true,
 
21
  "vocab_size": 119547
22
  }
 
1
  {
2
+ "_name_or_path": "old_models/distilbert-multilingual-nli-stsb-quora-ranking/0_Transformer",
3
  "activation": "gelu",
4
  "architectures": [
5
  "DistilBertModel"
 
19
  "seq_classif_dropout": 0.2,
20
  "sinusoidal_pos_embds": false,
21
  "tie_weights_": true,
22
+ "transformers_version": "4.7.0",
23
  "vocab_size": 119547
24
  }
config_sentence_transformers.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "2.0.0",
4
+ "transformers": "4.7.0",
5
+ "pytorch": "1.9.0+cu102"
6
+ }
7
+ }
modules.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ ]
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:66863c27df8fe2c66905a1c19be5848e4963f8b8d9193577b21e45b9449e453b
3
- size 538975987
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a9b43deff48b39cb5720f8ba664e69134b8cba3dac449d0feaf6e58d2513056
3
+ size 538971577
sentence_bert_config.json CHANGED
@@ -1,3 +1,4 @@
1
  {
2
- "max_seq_length": 128
 
3
  }
 
1
  {
2
+ "max_seq_length": 128,
3
+ "do_lower_case": false
4
  }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json CHANGED
@@ -1 +1 @@
1
- {"do_lower_case": false, "model_max_length": 512}
 
1
+ {"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": "old_models/distilbert-multilingual-nli-stsb-quora-ranking/0_Transformer/special_tokens_map.json", "name_or_path": "old_models/distilbert-multilingual-nli-stsb-quora-ranking/0_Transformer", "do_basic_tokenize": true, "never_split": null}