File size: 2,941 Bytes
d6960b0
 
 
 
 
d423311
 
d6960b0
 
 
 
 
 
 
d423311
 
 
 
 
 
 
 
 
 
 
 
 
ae4485d
d423311
 
ae4485d
d423311
 
ae4485d
d423311
 
ae4485d
d6960b0
 
 
 
 
 
 
d423311
d6960b0
ae4485d
 
 
 
 
d6960b0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ae4485d
d6960b0
 
 
 
 
ae4485d
 
 
 
 
 
 
 
 
 
d6960b0
 
 
 
 
 
 
 
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
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
---
license: apache-2.0
base_model: distilbert/distilroberta-base
tags:
- generated_from_trainer
datasets:
- lener_br
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: distilroberta-base-finetuned-ner-lenerBr
  results:
  - task:
      name: Token Classification
      type: token-classification
    dataset:
      name: lener_br
      type: lener_br
      config: lener_br
      split: validation
      args: lener_br
    metrics:
    - name: Precision
      type: precision
      value: 0.801254136909946
    - name: Recall
      type: recall
      value: 0.8429540040315191
    - name: F1
      type: f1
      value: 0.821575281300232
    - name: Accuracy
      type: accuracy
      value: 0.9685663231476382
---

<!-- 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. -->

# distilroberta-base-finetuned-ner-lenerBr

This model is a fine-tuned version of [distilbert/distilroberta-base](https://huggingface.co/distilbert/distilroberta-base) on the lener_br dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1550
- Precision: 0.8013
- Recall: 0.8430
- F1: 0.8216
- Accuracy: 0.9686

## 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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10

### Training results

| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| No log        | 1.0   | 490  | 0.1750          | 0.7347    | 0.6581 | 0.6942 | 0.9465   |
| 0.2808        | 2.0   | 980  | 0.1642          | 0.6954    | 0.7598 | 0.7262 | 0.9538   |
| 0.093         | 3.0   | 1470 | 0.1849          | 0.6708    | 0.7992 | 0.7294 | 0.9510   |
| 0.0557        | 4.0   | 1960 | 0.1403          | 0.7807    | 0.8345 | 0.8067 | 0.9668   |
| 0.0366        | 5.0   | 2450 | 0.1560          | 0.7775    | 0.8466 | 0.8106 | 0.9626   |
| 0.027         | 6.0   | 2940 | 0.1612          | 0.7342    | 0.8239 | 0.7764 | 0.9621   |
| 0.0204        | 7.0   | 3430 | 0.1632          | 0.7625    | 0.8356 | 0.7974 | 0.9644   |
| 0.015         | 8.0   | 3920 | 0.1748          | 0.7375    | 0.8442 | 0.7873 | 0.9615   |
| 0.0135        | 9.0   | 4410 | 0.1547          | 0.7930    | 0.8446 | 0.8180 | 0.9685   |
| 0.0101        | 10.0  | 4900 | 0.1550          | 0.8013    | 0.8430 | 0.8216 | 0.9686   |


### Framework versions

- Transformers 4.41.1
- Pytorch 2.1.2
- Datasets 2.19.1
- Tokenizers 0.19.1