Add model card
Browse files
README.md
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
|
5 |
+
tags:
|
6 |
+
- detoxification
|
7 |
+
|
8 |
+
licenses:
|
9 |
+
- cc-by-nc-sa
|
10 |
+
---
|
11 |
+
|
12 |
+
**Model Overview**
|
13 |
+
|
14 |
+
It is a SVD-compressed model of original BART-based detoxification model ]s-nlp/bart-base-detox][1].
|
15 |
+
|
16 |
+
**How to use**
|
17 |
+
|
18 |
+
```python
|
19 |
+
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
|
20 |
+
model = AutoModelForSeq2SeqLM \
|
21 |
+
.from_pretrained('s-nlp/bart-base-detox-svd', trust_remote_code=True)
|
22 |
+
tokenizer = AutoTokenizer.from_pretrained('facebook/bart-base')
|
23 |
+
|
24 |
+
toxics = ['that sick fuck is going to be out in 54 years.']
|
25 |
+
tokens = tokenizer(toxics)
|
26 |
+
tokens = model.generate(**tokens, num_return_sequences=1, do_sample=False,
|
27 |
+
temperature=1.0, repetition_penalty=10.0,
|
28 |
+
max_length=128, num_beams=5)
|
29 |
+
neutrals = tokenizer.decode(tokens[0, ...], skip_special_tokens=True)
|
30 |
+
print(neutrals) # stdout: She is going to be out in 54 years.
|
31 |
+
```
|
32 |
+
|
33 |
+
[1]: //s-nlp/bart-base-detox
|