pierluigic
commited on
Commit
•
732b48b
1
Parent(s):
9ce9005
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
---
|
4 |
+
# Cross-Encoder for Word Sense Relationships Classification
|
5 |
+
|
6 |
+
This model was trained on word sense relationships extracted by WordNet for the [semantic change type classification](https://github.com/ChangeIsKey/change-type-classification).
|
7 |
+
|
8 |
+
The model can be used to detect which kind of relatioships (among homonymy, antonymy, hypernonym, hyponymy, and co-hypnomy) intercur between word senses: Given a pair of word sense definitions, encode the query will all possible passages (e.g. retrieved with ElasticSearch). Then sort the passages in a decreasing order.
|
9 |
+
|
10 |
+
The training code is available here: [SBERT.net Training MS Marco](https://github.com/UKPLab/sentence-transformers/tree/master/examples/training/ms_marco)
|
11 |
+
|
12 |
+
|
13 |
+
## Usage with Transformers
|
14 |
+
|
15 |
+
```python
|
16 |
+
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
17 |
+
import torch
|
18 |
+
|
19 |
+
model = AutoModelForSequenceClassification.from_pretrained('model_name')
|
20 |
+
tokenizer = AutoTokenizer.from_pretrained('model_name')
|
21 |
+
|
22 |
+
features = tokenizer(['How many people live in Berlin?', 'How many people live in Berlin?'], ['Berlin has a population of 3,520,031 registered inhabitants in an area of 891.82 square kilometers.', 'New York City is famous for the Metropolitan Museum of Art.'], padding=True, truncation=True, return_tensors="pt")
|
23 |
+
|
24 |
+
model.eval()
|
25 |
+
with torch.no_grad():
|
26 |
+
scores = model(**features).logits
|
27 |
+
print(scores)
|
28 |
+
```
|
29 |
+
|
30 |
+
|
31 |
+
## Usage with SentenceTransformers
|
32 |
+
|
33 |
+
The usage becomes easier when you have [SentenceTransformers](https://www.sbert.net/) installed. Then, you can use the pre-trained models like this:
|
34 |
+
```python
|
35 |
+
from sentence_transformers import CrossEncoder
|
36 |
+
model = CrossEncoder('model_name', max_length=512)
|
37 |
+
labels = model.predict([('Query', 'Paragraph1'), ('Query', 'Paragraph2') , ('Query', 'Paragraph3')])
|
38 |
+
```
|
39 |
+
|
40 |
+
|
41 |
+
## Performance
|
42 |
+
In the following table, we provide various pre-trained Cross-Encoders together with their performance on the
|
43 |
+
|
44 |
+
![alt text](https://github.com/ChangeIsKey/change-type-classification/blob/main/lsc_ctd_benchmark_snippet_table.png "t")
|