File size: 2,515 Bytes
0ce4bcf
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c6b1148
 
 
 
 
0ce4bcf
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cc3130a
0ce4bcf
cc3130a
0ce4bcf
 
 
 
 
c6b1148
 
 
 
 
 
 
 
 
 
0ce4bcf
 
 
 
 
 
 
 
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
---
library_name: transformers
license: cc-by-nc-sa-4.0
base_model: microsoft/layoutlmv3-base
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: layoutlmv3-finetuned-invoice
  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. -->

# layoutlmv3-finetuned-invoice

This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0558
- Precision: 0.9435
- Recall: 0.9612
- F1: 0.9523
- Accuracy: 0.9858

## 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: 1e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- training_steps: 1000

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
|:-------------:|:------:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| No log        | 0.3195 | 100  | 0.0961          | 0.9121    | 0.9134 | 0.9128 | 0.9752   |
| No log        | 0.6390 | 200  | 0.0772          | 0.9120    | 0.9396 | 0.9256 | 0.9780   |
| No log        | 0.9585 | 300  | 0.0707          | 0.9272    | 0.9509 | 0.9389 | 0.9822   |
| No log        | 1.2780 | 400  | 0.0638          | 0.9202    | 0.9602 | 0.9398 | 0.9819   |
| 0.1131        | 1.5974 | 500  | 0.0631          | 0.9270    | 0.9582 | 0.9423 | 0.9829   |
| 0.1131        | 1.9169 | 600  | 0.0561          | 0.9331    | 0.9615 | 0.9471 | 0.9843   |
| 0.1131        | 2.2364 | 700  | 0.0651          | 0.9141    | 0.9720 | 0.9421 | 0.9824   |
| 0.1131        | 2.5559 | 800  | 0.0537          | 0.9515    | 0.9556 | 0.9535 | 0.9862   |
| 0.1131        | 2.8754 | 900  | 0.0556          | 0.9467    | 0.9582 | 0.9524 | 0.9860   |
| 0.0379        | 3.1949 | 1000 | 0.0558          | 0.9435    | 0.9612 | 0.9523 | 0.9858   |


### Framework versions

- Transformers 4.46.0.dev0
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.0