File size: 1,135 Bytes
6424b22
 
d1f2aa8
 
 
6424b22
 
 
 
 
d1f2aa8
 
6424b22
 
d1f2aa8
6424b22
d1f2aa8
6424b22
d1f2aa8
6424b22
 
 
 
 
 
 
d1f2aa8
6424b22
d1f2aa8
 
 
 
 
6424b22
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d1f2aa8
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
---
tags:
- summarization
language:
- it
metrics:
- rouge
model-index:
- name: summarization_mbart_mlsum
  results: []
datasets:
- ARTeLab/mlsum-it
---

# mbart_summarization_mlsum

This model is a fine-tuned version of [facebook/mbart-large-cc25](https://huggingface.co/facebook/mbart-large-cc25) on mlsum-it for Abstractive Summarization.

It achieves the following results:
- Loss: 3.3336
- Rouge1: 19.3489
- Rouge2: 6.4028
- Rougel: 16.3497
- Rougelsum: 16.5387
- Gen Len: 33.5945

## Usage 

```python
from transformers import MBartTokenizer, MBartForConditionalGeneration
tokenizer = MBartTokenizer.from_pretrained("ARTeLab/mbart-summarization-mlsum")
model = MBartForConditionalGeneration.from_pretrained("ARTeLab/mbart-summarization-mlsum")
```

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4.0

### Framework versions

- Transformers 4.15.0.dev0
- Pytorch 1.10.0+cu102
- Datasets 1.15.1
- Tokenizers 0.10.3