Pringled commited on
Commit
90d707e
1 Parent(s): 81ae9eb

Upload folder using huggingface_hub

Browse files
Files changed (2) hide show
  1. README.md +64 -0
  2. config.json +1 -1
README.md ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ model_name: M2V_base_multilingual
3
+ base_model: sentence-transformers/LaBSE
4
+ language: ['multilingual', 'af', 'sq', 'am', 'ar', 'hy', 'as', 'az', 'eu', 'be', 'bn', 'bs', 'bg', 'my', 'ca', 'ceb', 'zh', 'co', 'hr', 'cs', 'da', 'nl', 'en', 'eo', 'et', 'fi', 'fr', 'fy', 'gl', 'ka', 'de', 'el', 'gu', 'ht', 'ha', 'haw', 'he', 'hi', 'hmn', 'hu', 'is', 'ig', 'id', 'ga', 'it', 'ja', 'jv', 'kn', 'kk', 'km', 'rw', 'ko', 'ku', 'ky', 'lo', 'la', 'lv', 'lt', 'lb', 'mk', 'mg', 'ms', 'ml', 'mt', 'mi', 'mr', 'mn', 'ne', 'no', 'ny', 'or', 'fa', 'pl', 'pt', 'pa', 'ro', 'ru', 'sm', 'gd', 'sr', 'st', 'sn', 'si', 'sk', 'sl', 'so', 'es', 'su', 'sw', 'sv', 'tl', 'tg', 'ta', 'tt', 'te', 'th', 'bo', 'tr', 'tk', 'ug', 'uk', 'ur', 'uz', 'vi', 'cy', 'wo', 'xh', 'yi', 'yo', 'zu']
5
+ library_name: 'model2vec'
6
+ license: mit
7
+ tags: [embeddings, sentence-transformers, static-embeddings]
8
+ ---
9
+
10
+ # M2V_base_multilingual Model Card
11
+
12
+ Model2Vec distills a Sentence Transformer into a small, static model.
13
+ This model is ideal for applications requiring fast, lightweight embeddings.
14
+
15
+
16
+
17
+ ## Installation
18
+
19
+ Install model2vec using pip:
20
+ ```
21
+ pip install model2vec
22
+ ```
23
+
24
+ ## Usage
25
+ A StaticModel can be loaded using the `from_pretrained` method:
26
+ ```python
27
+ from model2vec import StaticModel
28
+ model = StaticModel.from_pretrained("minishlab/M2V_base_output")
29
+ embeddings = model.encode(["Example sentence"])
30
+ ```
31
+
32
+ Alternatively, you can distill your own model using the `distill` method:
33
+ ```python
34
+ from model2vec.distill import distill
35
+
36
+ # Choose a Sentence Transformer model
37
+ model_name = "BAAI/bge-base-en-v1.5"
38
+
39
+ # Distill the model
40
+ m2v_model = distill(model_name=model_name, pca_dims=256)
41
+
42
+ # Save the model
43
+ m2v_model.save_pretrained("m2v_model")
44
+ ```
45
+
46
+ ## How it works
47
+
48
+ Model2vec creates a small, fast, and powerful model that outperforms other static embedding models by a large margin on all tasks we could find, while being much faster to create than traditional static embedding models such as GloVe. Best of all, you don't need any data to distill a model using Model2Vec.
49
+
50
+ It works by passing a vocabulary through a sentence transformer model, then reducing the dimensionality of the resulting embeddings using PCA, and finally weighting the embeddings using zipf weighting. During inference, we simply take the mean of all token embeddings occurring in a sentence.
51
+
52
+ ## Citation
53
+
54
+ Please cite the [Model2Vec repository](https://github.com/MinishLab/model2vec) if you use this model in your work.
55
+
56
+ ## Additional Resources
57
+
58
+ - [Model2Vec Repo](https://github.com/MinishLab/model2vec)
59
+ - [Model2Vec Results](https://github.com/MinishLab/model2vec?tab=readme-ov-file#results)
60
+ - [Model2Vec Tutorials](https://github.com/MinishLab/model2vec/tree/main/tutorials)
61
+
62
+ ## Model Authors
63
+
64
+ Model2Vec was developed by the [Minish Lab](https://github.com/MinishLab) team consisting of Stephan Tulkens and Thomas van Dongen.
config.json CHANGED
@@ -1 +1 @@
1
- {"tokenizer_name": "sentence-transformers/LaBSE", "apply_pca": 256, "apply_zipf": true}
 
1
+ {"tokenizer_name": "sentence-transformers/LaBSE", "apply_pca": 256, "apply_zipf": true, "normalize": false}