File size: 2,473 Bytes
c5ef67f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bb490f7
c5ef67f
 
bb490f7
c5ef67f
 
bb490f7
c5ef67f
 
bb490f7
c5ef67f
 
 
 
 
 
 
bb490f7
c5ef67f
bb490f7
 
 
 
 
c5ef67f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bb490f7
c5ef67f
 
 
bb490f7
 
 
 
 
 
 
 
c5ef67f
 
 
 
 
 
 
 
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
---
tags:
- generated_from_trainer
datasets:
- data_registros_layoutv3
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: layoutlmv3-finetuned-registros_100
  results:
  - task:
      name: Token Classification
      type: token-classification
    dataset:
      name: data_registros_layoutv3
      type: data_registros_layoutv3
      config: default
      split: test
      args: default
    metrics:
    - name: Precision
      type: precision
      value: 0.9967585089141004
    - name: Recall
      type: recall
      value: 0.9951456310679612
    - name: F1
      type: f1
      value: 0.9959514170040485
    - name: Accuracy
      type: accuracy
      value: 0.999531542785759
---

<!-- 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. -->

# layoutlmv3-finetuned-registros_100

This model was trained from scratch on the data_registros_layoutv3 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0050
- Precision: 0.9968
- Recall: 0.9951
- F1: 0.9960
- Accuracy: 0.9995

## 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-06
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 600

### Training results

| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| No log        | 4.35  | 100  | 0.0106          | 0.9871    | 0.9935 | 0.9903 | 0.9991   |
| No log        | 8.7   | 200  | 0.0073          | 0.9984    | 0.9968 | 0.9976 | 0.9997   |
| No log        | 13.04 | 300  | 0.0061          | 0.9968    | 0.9968 | 0.9968 | 0.9997   |
| No log        | 17.39 | 400  | 0.0048          | 0.9968    | 0.9984 | 0.9976 | 0.9997   |
| 0.0109        | 21.74 | 500  | 0.0053          | 0.9968    | 0.9968 | 0.9968 | 0.9997   |
| 0.0109        | 26.09 | 600  | 0.0050          | 0.9968    | 0.9951 | 0.9960 | 0.9995   |


### Framework versions

- Transformers 4.29.2
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3