nreimers commited on
Commit
3d21d6b
1 Parent(s): a9eebee

Add new SentenceTransformer model.

Browse files
.gitattributes CHANGED
@@ -6,3 +6,6 @@
6
  *.tar.gz filter=lfs diff=lfs merge=lfs -text
7
  *.ot filter=lfs diff=lfs merge=lfs -text
8
  *.onnx filter=lfs diff=lfs merge=lfs -text
 
 
 
 
6
  *.tar.gz filter=lfs diff=lfs merge=lfs -text
7
  *.ot filter=lfs diff=lfs merge=lfs -text
8
  *.onnx filter=lfs diff=lfs merge=lfs -text
9
+ pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
10
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
11
+ .git/lfs/objects/48/31/48315809d75adfbf8e9922ee0cdaaae26b4f6680ba8595d7ae50d67de848c830 filter=lfs diff=lfs merge=lfs -text
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 CHANGED
@@ -1,18 +1,42 @@
1
  ---
 
2
  tags:
3
- - sentence_transformers
4
  - feature-extraction
 
 
5
  ---
6
- # Sentence Embeddings Models trained on Paraphrases
7
- This model is from the [sentence-transformers](https://github.com/UKPLab/sentence-transformers)-repository. It was trained on SNLI + MultiNLI and on STS benchmark dataset. Further details on SBERT can be found in the paper: [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://arxiv.org/abs/1908.10084)
8
 
9
- This model is multilingual version, it was trained on parallel data for 50+ languages.
10
 
11
- For more details, see: [SBERT.net - Pretrained Models](https://www.sbert.net/docs/pretrained_models.html)
12
 
13
- ## Usage (HuggingFace Models Repository)
14
 
15
- You can use the model directly from the model repository to compute sentence embeddings:
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
16
  ```python
17
  from transformers import AutoTokenizer, AutoModel
18
  import torch
@@ -22,62 +46,61 @@ import torch
22
  def mean_pooling(model_output, attention_mask):
23
  token_embeddings = model_output[0] #First element of model_output contains all token embeddings
24
  input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
25
- sum_embeddings = torch.sum(token_embeddings * input_mask_expanded, 1)
26
- sum_mask = torch.clamp(input_mask_expanded.sum(1), min=1e-9)
27
- return sum_embeddings / sum_mask
28
 
29
 
 
 
30
 
31
- #Sentences we want sentence embeddings for
32
- sentences = ['This framework generates embeddings for each input sentence',
33
- 'Sentences are passed as a list of string.',
34
- 'The quick brown fox jumps over the lazy dog.']
35
 
36
- #Load AutoModel from huggingface model repository
37
- tokenizer = AutoTokenizer.from_pretrained("model_name")
38
- model = AutoModel.from_pretrained("model_name")
39
 
40
- #Tokenize sentences
41
- encoded_input = tokenizer(sentences, padding=True, truncation=True, max_length=128, return_tensors='pt')
42
-
43
- #Compute token embeddings
44
  with torch.no_grad():
45
  model_output = model(**encoded_input)
46
 
47
- #Perform pooling. In this case, mean pooling
48
  sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
49
- ```
50
-
51
- ## Usage (Sentence-Transformers)
52
- Using this model becomes more convenient when you have [sentence-transformers](https://github.com/UKPLab/sentence-transformers) installed:
53
- ```
54
- pip install -U sentence-transformers
55
- ```
56
-
57
- Then you can use the model like this:
58
- ```python
59
- from sentence_transformers import SentenceTransformer
60
- model = SentenceTransformer('model_name')
61
- sentences = ['This framework generates embeddings for each input sentence',
62
- 'Sentences are passed as a list of string.',
63
- 'The quick brown fox jumps over the lazy dog.']
64
- sentence_embeddings = model.encode(sentences)
65
 
66
  print("Sentence embeddings:")
67
  print(sentence_embeddings)
68
  ```
69
 
70
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
71
  ## Citing & Authors
72
- If you find this model helpful, feel free to cite our publication [Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation](https://arxiv.org/abs/2004.09813):
73
- ```
74
- @inproceedings{reimers-2020-multilingual-sentence-bert,
75
- title = "Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation",
 
 
 
76
  author = "Reimers, Nils and Gurevych, Iryna",
77
- booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing",
78
  month = "11",
79
- year = "2020",
80
  publisher = "Association for Computational Linguistics",
81
- url = "https://arxiv.org/abs/2004.09813",
82
  }
83
  ```
 
1
  ---
2
+ pipeline_tag: sentence-similarity
3
  tags:
4
+ - sentence-transformers
5
  - feature-extraction
6
+ - sentence-similarity
7
+ - transformers
8
  ---
 
 
9
 
10
+ # sentence-transformers/stsb-xlm-r-multilingual
11
 
12
+ 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.
13
 
 
14
 
15
+
16
+ ## Usage (Sentence-Transformers)
17
+
18
+ Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:
19
+
20
+ ```
21
+ pip install -U sentence-transformers
22
+ ```
23
+
24
+ Then you can use the model like this:
25
+
26
+ ```python
27
+ from sentence_transformers import SentenceTransformer
28
+ sentences = ["This is an example sentence", "Each sentence is converted"]
29
+
30
+ model = SentenceTransformer('sentence-transformers/stsb-xlm-r-multilingual')
31
+ embeddings = model.encode(sentences)
32
+ print(embeddings)
33
+ ```
34
+
35
+
36
+
37
+ ## Usage (HuggingFace Transformers)
38
+ 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.
39
+
40
  ```python
41
  from transformers import AutoTokenizer, AutoModel
42
  import torch
 
46
  def mean_pooling(model_output, attention_mask):
47
  token_embeddings = model_output[0] #First element of model_output contains all token embeddings
48
  input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
49
+ return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
 
 
50
 
51
 
52
+ # Sentences we want sentence embeddings for
53
+ sentences = ['This is an example sentence', 'Each sentence is converted']
54
 
55
+ # Load model from HuggingFace Hub
56
+ tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/stsb-xlm-r-multilingual')
57
+ model = AutoModel.from_pretrained('sentence-transformers/stsb-xlm-r-multilingual')
 
58
 
59
+ # Tokenize sentences
60
+ encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
 
61
 
62
+ # Compute token embeddings
 
 
 
63
  with torch.no_grad():
64
  model_output = model(**encoded_input)
65
 
66
+ # Perform pooling. In this case, max pooling.
67
  sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
68
 
69
  print("Sentence embeddings:")
70
  print(sentence_embeddings)
71
  ```
72
 
73
 
74
+
75
+ ## Evaluation Results
76
+
77
+
78
+
79
+ For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name=sentence-transformers/stsb-xlm-r-multilingual)
80
+
81
+
82
+
83
+ ## Full Model Architecture
84
+ ```
85
+ SentenceTransformer(
86
+ (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
87
+ (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})
88
+ )
89
+ ```
90
+
91
  ## Citing & Authors
92
+
93
+ This model was trained by [sentence-transformers](https://www.sbert.net/).
94
+
95
+ 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):
96
+ ```bibtex
97
+ @inproceedings{reimers-2019-sentence-bert,
98
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
99
  author = "Reimers, Nils and Gurevych, Iryna",
100
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
101
  month = "11",
102
+ year = "2019",
103
  publisher = "Association for Computational Linguistics",
104
+ url = "http://arxiv.org/abs/1908.10084",
105
  }
106
  ```
config.json CHANGED
@@ -1,4 +1,5 @@
1
  {
 
2
  "architectures": [
3
  "XLMRobertaModel"
4
  ],
@@ -18,6 +19,9 @@
18
  "num_hidden_layers": 12,
19
  "output_past": true,
20
  "pad_token_id": 1,
 
 
21
  "type_vocab_size": 1,
 
22
  "vocab_size": 250002
23
  }
 
1
  {
2
+ "_name_or_path": "old_models/stsb-xlm-r-multilingual/0_Transformer",
3
  "architectures": [
4
  "XLMRobertaModel"
5
  ],
 
19
  "num_hidden_layers": 12,
20
  "output_past": true,
21
  "pad_token_id": 1,
22
+ "position_embedding_type": "absolute",
23
+ "transformers_version": "4.7.0",
24
  "type_vocab_size": 1,
25
+ "use_cache": true,
26
  "vocab_size": 250002
27
  }
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 sha256:48315809d75adfbf8e9922ee0cdaaae26b4f6680ba8595d7ae50d67de848c830
3
- size 1112256686
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e6e54271c84414d1031eec8424e9698d19ff3dab29cdd652ecf4b360968b1de
3
+ size 1112253233
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
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96ef91dd5d1c3b157a0437e85ff1f29d0eb3cf32fd40f863dab45baa5e839fad
3
+ size 9096735
tokenizer_config.json CHANGED
@@ -1 +1 @@
1
- {"model_max_length": 512, "special_tokens_map_file": "output/xlm-r-nli-stsb-40langs/0_Transformer/special_tokens_map.json", "full_tokenizer_file": null}
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "model_max_length": 512, "special_tokens_map_file": "output/xlm-r-nli-stsb-40langs/0_Transformer/special_tokens_map.json", "full_tokenizer_file": null, "name_or_path": "old_models/stsb-xlm-r-multilingual/0_Transformer", "sp_model_kwargs": {}}