File size: 8,025 Bytes
41ab1f5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ffd9c00
 
 
 
 
 
 
 
41ab1f5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ffd9c00
41ab1f5
 
 
 
ffd9c00
 
 
 
 
 
 
 
 
 
 
 
 
 
41ab1f5
 
 
 
 
 
 
 
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
---
license: mit
base_model: microsoft/layoutlm-base-uncased
tags:
- generated_from_trainer
datasets:
- funsd
model-index:
- name: layoutlm-funsd2
  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. -->

# layoutlm-funsd2

This model is a fine-tuned version of [microsoft/layoutlm-base-uncased](https://huggingface.co/microsoft/layoutlm-base-uncased) on the funsd dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6614
- Answer: {'precision': 0.6683778234086243, 'recall': 0.8046971569839307, 'f1': 0.7302299495232752, 'number': 809}
- Header: {'precision': 0.3130434782608696, 'recall': 0.3025210084033613, 'f1': 0.3076923076923077, 'number': 119}
- Question: {'precision': 0.7667814113597247, 'recall': 0.8366197183098592, 'f1': 0.8001796138302649, 'number': 1065}
- Overall Precision: 0.7010
- Overall Recall: 0.7918
- Overall F1: 0.7436
- Overall Accuracy: 0.8029

## 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: 3e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 12
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Answer                                                                                                         | Header                                                                                                       | Question                                                                                                    | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------:|:-----------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
| 1.8071        | 1.0   | 10   | 1.5850          | {'precision': 0.011918951132300357, 'recall': 0.012360939431396786, 'f1': 0.012135922330097087, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}                                                  | {'precision': 0.17111459968602827, 'recall': 0.10234741784037558, 'f1': 0.1280846063454759, 'number': 1065} | 0.0806            | 0.0597         | 0.0686     | 0.3795           |
| 1.4934        | 2.0   | 20   | 1.2707          | {'precision': 0.09924812030075188, 'recall': 0.0815822002472188, 'f1': 0.08955223880597016, 'number': 809}     | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}                                                  | {'precision': 0.4546952224052718, 'recall': 0.5183098591549296, 'f1': 0.484422992540588, 'number': 1065}    | 0.3289            | 0.3101         | 0.3192     | 0.5753           |
| 1.1823        | 3.0   | 30   | 0.9970          | {'precision': 0.4033214709371293, 'recall': 0.42027194066749074, 'f1': 0.4116222760290557, 'number': 809}      | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}                                                  | {'precision': 0.5919540229885057, 'recall': 0.6769953051643193, 'f1': 0.6316250547525186, 'number': 1065}   | 0.5106            | 0.5324         | 0.5212     | 0.6915           |
| 0.9185        | 4.0   | 40   | 0.8213          | {'precision': 0.6075156576200418, 'recall': 0.7194066749072929, 'f1': 0.6587436332767402, 'number': 809}       | {'precision': 0.05128205128205128, 'recall': 0.01680672268907563, 'f1': 0.025316455696202535, 'number': 119} | {'precision': 0.6559048428207307, 'recall': 0.7248826291079812, 'f1': 0.6886708296164139, 'number': 1065}   | 0.6237            | 0.6804         | 0.6508     | 0.7467           |
| 0.7233        | 5.0   | 50   | 0.7353          | {'precision': 0.638974358974359, 'recall': 0.7700865265760197, 'f1': 0.6984304932735426, 'number': 809}        | {'precision': 0.22093023255813954, 'recall': 0.15966386554621848, 'f1': 0.18536585365853656, 'number': 119}  | {'precision': 0.6809716599190283, 'recall': 0.7896713615023474, 'f1': 0.731304347826087, 'number': 1065}    | 0.6459            | 0.7441         | 0.6915     | 0.7794           |
| 0.6262        | 6.0   | 60   | 0.7036          | {'precision': 0.632512315270936, 'recall': 0.7935723114956736, 'f1': 0.7039473684210525, 'number': 809}        | {'precision': 0.24324324324324326, 'recall': 0.15126050420168066, 'f1': 0.18652849740932642, 'number': 119}  | {'precision': 0.7235345581802275, 'recall': 0.7765258215962442, 'f1': 0.7490942028985508, 'number': 1065}   | 0.6662            | 0.7461         | 0.7039     | 0.7818           |
| 0.5552        | 7.0   | 70   | 0.6694          | {'precision': 0.6639089968976215, 'recall': 0.7935723114956736, 'f1': 0.722972972972973, 'number': 809}        | {'precision': 0.24770642201834864, 'recall': 0.226890756302521, 'f1': 0.23684210526315788, 'number': 119}    | {'precision': 0.730999146029035, 'recall': 0.8037558685446009, 'f1': 0.7656529516994633, 'number': 1065}    | 0.6787            | 0.7652         | 0.7193     | 0.7913           |
| 0.5016        | 8.0   | 80   | 0.6598          | {'precision': 0.6592517694641051, 'recall': 0.8059332509270705, 'f1': 0.7252502780867631, 'number': 809}       | {'precision': 0.24324324324324326, 'recall': 0.226890756302521, 'f1': 0.23478260869565218, 'number': 119}    | {'precision': 0.7482817869415808, 'recall': 0.8178403755868544, 'f1': 0.781516375056079, 'number': 1065}    | 0.6846            | 0.7777         | 0.7282     | 0.7931           |
| 0.4496        | 9.0   | 90   | 0.6561          | {'precision': 0.6663265306122449, 'recall': 0.8071693448702101, 'f1': 0.7300167691447736, 'number': 809}       | {'precision': 0.2743362831858407, 'recall': 0.2605042016806723, 'f1': 0.26724137931034486, 'number': 119}    | {'precision': 0.7584708948740226, 'recall': 0.819718309859155, 'f1': 0.7879061371841156, 'number': 1065}    | 0.6939            | 0.7812         | 0.7350     | 0.7982           |
| 0.4481        | 10.0  | 100  | 0.6633          | {'precision': 0.6711340206185566, 'recall': 0.8046971569839307, 'f1': 0.7318718381112984, 'number': 809}       | {'precision': 0.29357798165137616, 'recall': 0.2689075630252101, 'f1': 0.28070175438596495, 'number': 119}   | {'precision': 0.7640350877192983, 'recall': 0.8178403755868544, 'f1': 0.7900226757369614, 'number': 1065}   | 0.7003            | 0.7797         | 0.7379     | 0.7987           |
| 0.4012        | 11.0  | 110  | 0.6624          | {'precision': 0.6625766871165644, 'recall': 0.8009888751545118, 'f1': 0.7252378287632905, 'number': 809}       | {'precision': 0.3333333333333333, 'recall': 0.3025210084033613, 'f1': 0.3171806167400881, 'number': 119}     | {'precision': 0.7696969696969697, 'recall': 0.8347417840375587, 'f1': 0.8009009009009008, 'number': 1065}   | 0.7019            | 0.7893         | 0.7430     | 0.8074           |
| 0.4065        | 12.0  | 120  | 0.6614          | {'precision': 0.6683778234086243, 'recall': 0.8046971569839307, 'f1': 0.7302299495232752, 'number': 809}       | {'precision': 0.3130434782608696, 'recall': 0.3025210084033613, 'f1': 0.3076923076923077, 'number': 119}     | {'precision': 0.7667814113597247, 'recall': 0.8366197183098592, 'f1': 0.8001796138302649, 'number': 1065}   | 0.7010            | 0.7918         | 0.7436     | 0.8029           |


### Framework versions

- Transformers 4.41.2
- Pytorch 2.3.1+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1