File size: 1,703 Bytes
60727ef
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: mit
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: punctuation-nilc-t5-base
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# punctuation-nilc-t5-base

This model is a fine-tuned version of [unicamp-dl/ptt5-base-portuguese-vocab](https://huggingface.co/unicamp-dl/ptt5-base-portuguese-vocab) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0432
- Bleu: 26.0973
- Gen Len: 18.8694

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5.0

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Bleu    | Gen Len |
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
| 0.0544        | 1.0   | 4686  | 0.0442          | 25.989  | 18.8655 |
| 0.0367        | 2.0   | 9372  | 0.0371          | 25.9358 | 18.8713 |
| 0.0222        | 3.0   | 14058 | 0.0374          | 25.8976 | 18.8694 |
| 0.0152        | 4.0   | 18744 | 0.0409          | 26.1575 | 18.8694 |
| 0.0147        | 5.0   | 23430 | 0.0432          | 26.0973 | 18.8694 |


### Framework versions

- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.2