Update README.md
Browse files
README.md
CHANGED
@@ -7,8 +7,25 @@ tags:
|
|
7 |
pipeline_tag: text-classification
|
8 |
---
|
9 |
|
|
|
|
|
10 |
# multi-label_8_sent_v3_1
|
11 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
This is a [SetFit model](https://github.com/huggingface/setfit) that can be used for text classification. The model has been trained using an efficient few-shot learning technique that involves:
|
13 |
|
14 |
1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
|
@@ -30,7 +47,7 @@ from setfit import SetFitModel
|
|
30 |
# Download from Hub and run inference
|
31 |
model = SetFitModel.from_pretrained("multi-label_8_sent_v3_1")
|
32 |
# Run inference
|
33 |
-
preds = model(["
|
34 |
```
|
35 |
|
36 |
## BibTeX entry and citation info
|
|
|
7 |
pipeline_tag: text-classification
|
8 |
---
|
9 |
|
10 |
+
|
11 |
+
|
12 |
# multi-label_8_sent_v3_1
|
13 |
|
14 |
+
This experimental multi-label model has been finetuned on NSS data to classify comments into 8 topic classes:
|
15 |
+
|
16 |
+
1. Teaching & Learning
|
17 |
+
2. Support
|
18 |
+
3. Communication
|
19 |
+
4. Organisation and timetable
|
20 |
+
5. Assessment and feedback
|
21 |
+
6. Career and placement
|
22 |
+
7. Health, wellbeing and social life
|
23 |
+
8. Facilities and technology
|
24 |
+
|
25 |
+
Inference works best at sentence-level.
|
26 |
+
|
27 |
+
# SetFit
|
28 |
+
|
29 |
This is a [SetFit model](https://github.com/huggingface/setfit) that can be used for text classification. The model has been trained using an efficient few-shot learning technique that involves:
|
30 |
|
31 |
1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
|
|
|
47 |
# Download from Hub and run inference
|
48 |
model = SetFitModel.from_pretrained("multi-label_8_sent_v3_1")
|
49 |
# Run inference
|
50 |
+
preds = model(["My lecturers were excellent!", "I wish we'd had more support when it came to assignments."])
|
51 |
```
|
52 |
|
53 |
## BibTeX entry and citation info
|