File size: 1,988 Bytes
e895756
 
 
 
 
 
 
 
 
 
 
ebb63b6
e895756
a2ce5dc
 
 
 
 
 
 
 
c9f8674
a2ce5dc
 
 
 
e895756
 
 
 
f437c48
e895756
a2ce5dc
 
 
 
 
 
 
 
 
 
 
 
e895756
 
a2ce5dc
 
e895756
 
 
 
 
 
a2ce5dc
e895756
 
 
 
a0977b7
a2ce5dc
 
e414d9b
f437c48
a2ce5dc
 
a0977b7
a2ce5dc
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
---
license: apache-2.0
metrics:
- accuracy
- f1
pipeline_tag: text-classification
tags:
- irony
language:
- it
---
# Irony at aequa-tech

## Model Description

- **Developed by:** [aequa-tech](https://aequa-tech.com/)
- **Funded by:** [NGI-Search](https://www.ngi.eu/ngi-projects/ngi-search/)
- **Language(s) (NLP):** Italian
- **License:** apache-2.0
- **Finetuned from model:** [AlBERTo](https://huggingface.co/m-polignano-uniba/bert_uncased_L-12_H-768_A-12_italian_alberto)

This model is a fine-tuned version of [AlBERTo](https://huggingface.co/m-polignano-uniba/bert_uncased_L-12_H-768_A-12_italian_alberto) Italian model on **irony detection**

# Training Details

## Training Data

- [IronITA 2018](https://live.european-language-grid.eu/catalogue/corpus/7372)
- [Sarcastic Hate Speech dataset](https://github.com/simonasnow/Sarcastic-Hate-Speech)
- SENTIPOLC [2014](https://live.european-language-grid.eu/catalogue/corpus/7480)/[2016](https://live.european-language-grid.eu/catalogue/corpus/7479)
- [Debunker-Assistant corpus](https://github.com/AequaTech/DebunkerAssistant/tree/main/evaluation/training_datasets)

## Training Hyperparameters

- learning_rate: 2e-5
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam


# Evaluation

## Testing Data
It was tested on IronITA test set obtaining the following results:

## Metrics and Results

- macro F1: 0.79
- accuracy: 0.79
- precision of positive class: 0.77
- recall of positive class: 0.84
- F1 of positive class: 0.80

# Framework versions

- Transformers 4.30.2
- Pytorch 2.1.2
- Datasets 2.19.0
- Accelerate 0.30.0

# How to use this model:
```Python
model = AutoModelForSequenceClassification.from_pretrained('aequa-tech/irony-it',num_labels=2) 
tokenizer = AutoTokenizer.from_pretrained("m-polignano-uniba/bert_uncased_L-12_H-768_A-12_italian_alb3rt0") 
classifier = pipeline("text-classification", model=model, tokenizer=tokenizer)
classifier("Prendi una gioia. Ora posala, che non è tua.")
```