nreimers commited on
Commit
7dbfdec
1 Parent(s): ce59f97

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 +1,131 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ ---
34
+
35
+ # sentence-transformers/distilbert-base-nli-stsb-quora-ranking
36
+
37
+ 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.
38
+
39
+
40
+
41
+ ## Usage (Sentence-Transformers)
42
+
43
+ Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:
44
+
45
+ ```
46
+ pip install -U sentence-transformers
47
+ ```
48
+
49
+ Then you can use the model like this:
50
+
51
+ ```python
52
+ from sentence_transformers import SentenceTransformer
53
+ sentences = ["This is an example sentence", "Each sentence is converted"]
54
+
55
+ model = SentenceTransformer('sentence-transformers/distilbert-base-nli-stsb-quora-ranking')
56
+ embeddings = model.encode(sentences)
57
+ print(embeddings)
58
+ ```
59
+
60
+
61
+
62
+ ## Usage (HuggingFace Transformers)
63
+ 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.
64
+
65
+ ```python
66
+ from transformers import AutoTokenizer, AutoModel
67
+ import torch
68
+
69
+
70
+ #Mean Pooling - Take attention mask into account for correct averaging
71
+ def mean_pooling(model_output, attention_mask):
72
+ token_embeddings = model_output[0] #First element of model_output contains all token embeddings
73
+ input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
74
+ return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
75
+
76
+
77
+ # Sentences we want sentence embeddings for
78
+ sentences = ['This is an example sentence', 'Each sentence is converted']
79
+
80
+ # Load model from HuggingFace Hub
81
+ tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/distilbert-base-nli-stsb-quora-ranking')
82
+ model = AutoModel.from_pretrained('sentence-transformers/distilbert-base-nli-stsb-quora-ranking')
83
+
84
+ # Tokenize sentences
85
+ encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
86
+
87
+ # Compute token embeddings
88
+ with torch.no_grad():
89
+ model_output = model(**encoded_input)
90
+
91
+ # Perform pooling. In this case, max pooling.
92
+ sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
93
+
94
+ print("Sentence embeddings:")
95
+ print(sentence_embeddings)
96
+ ```
97
+
98
+
99
+
100
+ ## Evaluation Results
101
+
102
+
103
+
104
+ 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-base-nli-stsb-quora-ranking)
105
+
106
+
107
+
108
+ ## Full Model Architecture
109
+ ```
110
+ SentenceTransformer(
111
+ (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: DistilBertModel
112
+ (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})
113
+ )
114
+ ```
115
+
116
+ ## Citing & Authors
117
+
118
+ This model was trained by [sentence-transformers](https://www.sbert.net/).
119
+
120
+ 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):
121
+ ```bibtex
122
+ @inproceedings{reimers-2019-sentence-bert,
123
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
124
+ author = "Reimers, Nils and Gurevych, Iryna",
125
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
126
+ month = "11",
127
+ year = "2019",
128
+ publisher = "Association for Computational Linguistics",
129
+ url = "http://arxiv.org/abs/1908.10084",
130
+ }
131
+ ```
config.json CHANGED
@@ -1,4 +1,5 @@
1
  {
 
2
  "activation": "gelu",
3
  "architectures": [
4
  "DistilBertModel"
@@ -17,5 +18,6 @@
17
  "seq_classif_dropout": 0.2,
18
  "sinusoidal_pos_embds": false,
19
  "tie_weights_": true,
 
20
  "vocab_size": 30522
21
  }
 
1
  {
2
+ "_name_or_path": "old_models/distilbert-base-nli-stsb-quora-ranking/0_Transformer",
3
  "activation": "gelu",
4
  "architectures": [
5
  "DistilBertModel"
 
18
  "seq_classif_dropout": 0.2,
19
  "sinusoidal_pos_embds": false,
20
  "tie_weights_": true,
21
+ "transformers_version": "4.7.0",
22
  "vocab_size": 30522
23
  }
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 CHANGED
@@ -2,7 +2,7 @@
2
  {
3
  "idx": 0,
4
  "name": "0",
5
- "path": "0_Transformer",
6
  "type": "sentence_transformers.models.Transformer"
7
  },
8
  {
 
2
  {
3
  "idx": 0,
4
  "name": "0",
5
+ "path": "",
6
  "type": "sentence_transformers.models.Transformer"
7
  },
8
  {
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:68d315c1638882d9c6cd8901ed497a24de830506773274f60f477620002e113d
3
- size 265473819
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d249f8d48aa2bf254c4ac1672f030873eff02e67aaeb0b9e9bdedb79cd4468cd
3
+ size 265486777
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": true, "model_max_length": 512, "special_tokens_map_file": "output/training_nli_distilbert-base-uncased-2020-07-22_10-20-15/0_Transformer/special_tokens_map.json", "full_tokenizer_file": null}
 
1
+ {"do_lower_case": true, "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": "output/training_nli_distilbert-base-uncased-2020-07-22_10-20-15/0_Transformer/special_tokens_map.json", "full_tokenizer_file": null, "name_or_path": "old_models/distilbert-base-nli-stsb-quora-ranking/0_Transformer", "do_basic_tokenize": true, "never_split": null}