Update README.md
Browse files
README.md
CHANGED
@@ -1,63 +1,64 @@
|
|
1 |
---
|
2 |
pipeline_tag: sentence-similarity
|
|
|
3 |
tags:
|
4 |
- sentence-transformers
|
5 |
- feature-extraction
|
6 |
- sentence-similarity
|
7 |
- transformers
|
|
|
|
|
8 |
---
|
9 |
|
10 |
-
#
|
11 |
|
12 |
-
This is a
|
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 |
-
|
|
|
|
|
25 |
|
|
|
26 |
```python
|
27 |
from sentence_transformers import SentenceTransformer
|
28 |
-
|
|
|
|
|
|
|
|
|
29 |
|
30 |
model = SentenceTransformer('clips/mfaq')
|
31 |
-
embeddings = model.encode(
|
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
|
43 |
|
44 |
-
|
45 |
-
#Mean Pooling - Take attention mask into account for correct averaging
|
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('clips/mfaq')
|
57 |
model = AutoModel.from_pretrained('clips/mfaq')
|
58 |
|
59 |
# Tokenize sentences
|
60 |
-
encoded_input = tokenizer(
|
61 |
|
62 |
# Compute token embeddings
|
63 |
with torch.no_grad():
|
@@ -65,29 +66,4 @@ with torch.no_grad():
|
|
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 |
-
<!--- Describe how your model was evaluated -->
|
78 |
-
|
79 |
-
For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name=clips/mfaq)
|
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 |
-
<!--- Describe where people can find more information -->
|
|
|
1 |
---
|
2 |
pipeline_tag: sentence-similarity
|
3 |
+
license: apache-2.0
|
4 |
tags:
|
5 |
- sentence-transformers
|
6 |
- feature-extraction
|
7 |
- sentence-similarity
|
8 |
- transformers
|
9 |
+
datasets:
|
10 |
+
- clips/mfaq
|
11 |
---
|
12 |
|
13 |
+
# MFAQ
|
14 |
|
15 |
+
This is a FAQ retrieval model, it ranks potential answers according to a given question. It was trained using the [MFAQ dataset](https://huggingface.co/datasets/clips/mfaq).
|
16 |
|
17 |
+
## Installation
|
|
|
|
|
|
|
|
|
18 |
|
19 |
```
|
20 |
pip install -U sentence-transformers
|
21 |
```
|
22 |
|
23 |
+
## Usage
|
24 |
+
You can use MFAQ with sentence-transformers or directly with a HuggingFace model.
|
25 |
+
In both cases, questions need to be prepended with `<Q>`, and answers with `<A>`.
|
26 |
|
27 |
+
#### Sentence Transformers
|
28 |
```python
|
29 |
from sentence_transformers import SentenceTransformer
|
30 |
+
|
31 |
+
question = "<Q>How many models can I host on HuggingFace?"
|
32 |
+
answer_1 = "<A>All plans come with unlimited private models and datasets."
|
33 |
+
answer_2 = "<A>AutoNLP is an automatic way to train and deploy state-of-the-art NLP models, seamlessly integrated with the Hugging Face ecosystem."
|
34 |
+
answer_3 = "<A>Based on how much training data and model variants are created, we send you a compute cost and payment link - as low as $10 per job."
|
35 |
|
36 |
model = SentenceTransformer('clips/mfaq')
|
37 |
+
embeddings = model.encode([question, answer_1, answer_3, answer_3])
|
38 |
print(embeddings)
|
39 |
```
|
40 |
|
41 |
+
#### HuggingFace Transfoormers
|
|
|
|
|
|
|
42 |
|
43 |
```python
|
44 |
from transformers import AutoTokenizer, AutoModel
|
45 |
import torch
|
46 |
|
|
|
|
|
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 |
return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
|
51 |
|
52 |
+
question = "<Q>How many models can I host on HuggingFace?"
|
53 |
+
answer_1 = "<A>All plans come with unlimited private models and datasets."
|
54 |
+
answer_2 = "<A>AutoNLP is an automatic way to train and deploy state-of-the-art NLP models, seamlessly integrated with the Hugging Face ecosystem."
|
55 |
+
answer_3 = "<A>Based on how much training data and model variants are created, we send you a compute cost and payment link - as low as $10 per job."
|
56 |
|
|
|
|
|
|
|
|
|
57 |
tokenizer = AutoTokenizer.from_pretrained('clips/mfaq')
|
58 |
model = AutoModel.from_pretrained('clips/mfaq')
|
59 |
|
60 |
# Tokenize sentences
|
61 |
+
encoded_input = tokenizer([question, answer_1, answer_3, answer_3], padding=True, truncation=True, return_tensors='pt')
|
62 |
|
63 |
# Compute token embeddings
|
64 |
with torch.no_grad():
|
|
|
66 |
|
67 |
# Perform pooling. In this case, max pooling.
|
68 |
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
|
69 |
+
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|