Commit
•
fa3cef4
1
Parent(s):
f8d1ea5
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,81 @@
|
|
1 |
---
|
2 |
license: mit
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: mit
|
3 |
---
|
4 |
+
|
5 |
+
---
|
6 |
+
tags:
|
7 |
+
- sentence-transformers
|
8 |
+
- feature-extraction
|
9 |
+
---
|
10 |
+
# Name of Model
|
11 |
+
|
12 |
+
<!--- Describe your model here -->
|
13 |
+
|
14 |
+
## Model Description
|
15 |
+
The model consists of the following layers:
|
16 |
+
|
17 |
+
(0) Base Transformer Type: BAAI/bge-small-en-v1.5
|
18 |
+
|
19 |
+
(1) mean Pooling
|
20 |
+
|
21 |
+
|
22 |
+
## Usage (Sentence-Transformers)
|
23 |
+
|
24 |
+
Using this model becomes more convenient when you have [sentence-transformers](https://github.com/UKPLab/sentence-transformers) installed:
|
25 |
+
|
26 |
+
```
|
27 |
+
pip install -U sentence-transformers
|
28 |
+
```
|
29 |
+
|
30 |
+
Then you can use the model like this:
|
31 |
+
|
32 |
+
```python
|
33 |
+
from sentence_transformers import SentenceTransformer
|
34 |
+
sentences = ["This is an example sentence"]
|
35 |
+
model = SentenceTransformer('model_name')
|
36 |
+
embeddings = model.encode(sentences)
|
37 |
+
print(embeddings)
|
38 |
+
```
|
39 |
+
|
40 |
+
|
41 |
+
## Usage (HuggingFace Transformers)
|
42 |
+
|
43 |
+
```python
|
44 |
+
from transformers import AutoTokenizer, AutoModel
|
45 |
+
import torch
|
46 |
+
#Mean Pooling - Take attention mask into account for correct averaging
|
47 |
+
def mean_pooling(model_output, attention_mask):
|
48 |
+
token_embeddings = model_output[0] #First element of model_output contains all token embeddings
|
49 |
+
input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
|
50 |
+
sum_embeddings = torch.sum(token_embeddings * input_mask_expanded, 1)
|
51 |
+
sum_mask = torch.clamp(input_mask_expanded.sum(1), min=1e-9)
|
52 |
+
return sum_embeddings / sum_mask
|
53 |
+
# Sentences we want sentence embeddings for
|
54 |
+
sentences = ['This is an example sentence']
|
55 |
+
# Load model from HuggingFace Hub
|
56 |
+
tokenizer = AutoTokenizer.from_pretrained('model_name')
|
57 |
+
model = AutoModel.from_pretrained('model_name')
|
58 |
+
# Tokenize sentences
|
59 |
+
encoded_input = tokenizer(sentences, padding=True, truncation=True, max_length=128, return_tensors='pt')
|
60 |
+
# Compute token embeddings
|
61 |
+
with torch.no_grad():
|
62 |
+
model_output = model(**encoded_input)
|
63 |
+
# Perform pooling. In this case, max pooling.
|
64 |
+
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
|
65 |
+
print("Sentence embeddings:")
|
66 |
+
print(sentence_embeddings)
|
67 |
+
```
|
68 |
+
|
69 |
+
|
70 |
+
|
71 |
+
## Training Procedure
|
72 |
+
|
73 |
+
<!--- Describe how your model was trained -->
|
74 |
+
|
75 |
+
## Evaluation Results
|
76 |
+
|
77 |
+
<!--- Describe how your model was evaluated -->
|
78 |
+
|
79 |
+
## Citing & Authors
|
80 |
+
|
81 |
+
<!--- Describe where people can find more information -->
|