File size: 1,187 Bytes
72c61d6
 
 
96b09a2
 
72c61d6
 
 
 
 
96b09a2
 
 
 
 
72c61d6
 
2b53f5a
72c61d6
63978b0
72c61d6
 
 
 
 
 
 
 
 
 
 
63978b0
72c61d6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
96b09a2
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
---
library_name: transformers
license: mit
base_model:
- neuralmind/bert-base-portuguese-cased
tags:
- generated_from_trainer
model-index:
- name: congretimbau3
  results: []
datasets:
- belisards/ementas_senado_1946_2024
- belisards/ementas_camarabr_1934_2024
language:
- pt
---

# congretimbau

This model is a fine-tuned version of [BERTimbau](https://huggingface.co/neuralmind/bert-base-portuguese-cased) on a dataset with bills of Brazilian law proposals.
It achieves the following results on the evaluation set:
- eval_loss: 0.4885
- eval_runtime: 798.5704
- eval_samples_per_second: 169.279
- eval_steps_per_second: 1.324
- epoch: 2.3669
- step: 10000


## Training and evaluation data

Data from the Chamber of Deputies and the Federal Senate.

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 300
- num_epochs: 10

### Framework versions

- Transformers 4.45.1
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.0