Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- it
|
4 |
+
---
|
5 |
+
## GPT-ita-fdi_lega🇮🇹
|
6 |
+
|
7 |
+
|
8 |
+
Finetune of an Italian version of gpt-2 ((GePpeTto)[https://huggingface.co/LorenzoDeMattei/GePpeTto]) trained on tweets of politicians from the far right Italian parties FDI and Lega.
|
9 |
+
|
10 |
+
|
11 |
+
## Finetuning corpus
|
12 |
+
|
13 |
+
The model was finetuned over a private dataset of tweets from italian politicians. The tweets were collected between 2021 and 2022 from the Twitter accounts of all the "FDI" and "Lega" members of the Italian Parliament.
|
14 |
+
In the end, the finetuning was conducted over a corpus of ~40K tweets
|
15 |
+
|
16 |
+
## Uses
|
17 |
+
|
18 |
+
By giving the model a few Italian words to start from, the model can generate a tweet in the style of far right Italian politicians. Try it out (here)[https://huggingface.co/spaces/ruggsea/demo_gpt-ita-fdi_lega]
|
19 |
+
|
20 |
+
|
21 |
+
## Bias, Risks, and Limitations
|
22 |
+
|
23 |
+
Compared to the base italian gpt-2 model, this model could generate more hateful or toxic content and exhibit bias, in line with the training corpus.
|
24 |
+
|
25 |
+
### Recommendations
|
26 |
+
|
27 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
28 |
+
|
29 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
30 |
+
|
31 |
+
## How to Get Started with the Model
|
32 |
+
|
33 |
+
Use the code below to get started with the model.
|
34 |
+
|
35 |
+
|
36 |
+
'''
|
37 |
+
from transformers import GPT2Tokenizer, GPT2Model
|
38 |
+
|
39 |
+
model = GPT2Model.from_pretrained('ruggsea/gpt-ita-fdi_lega')
|
40 |
+
tokenizer = GPT2Tokenizer.from_pretrained(
|
41 |
+
'ruggsea/gpt-ita-fdi_lega',
|
42 |
+
)
|
43 |
+
'''
|