Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,76 @@
|
|
1 |
---
|
2 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
tags:
|
3 |
+
- generated_from_trainer
|
4 |
+
datasets:
|
5 |
+
- nielsr/funsd-layoutlmv3
|
6 |
+
metrics:
|
7 |
+
- precision
|
8 |
+
- recall
|
9 |
+
- f1
|
10 |
+
- accuracy
|
11 |
+
base_model: microsoft/layoutlmv3-base
|
12 |
+
model-index:
|
13 |
+
- name: layoutlmv3-finetuned-funsd
|
14 |
+
results:
|
15 |
+
- task:
|
16 |
+
type: token-classification
|
17 |
+
name: Token Classification
|
18 |
+
dataset:
|
19 |
+
name: nielsr/funsd-layoutlmv3
|
20 |
+
type: nielsr/funsd-layoutlmv3
|
21 |
+
args: funsd
|
22 |
+
metrics:
|
23 |
+
- type: precision
|
24 |
+
value: 0.9026198714780029
|
25 |
+
name: Precision
|
26 |
+
- type: recall
|
27 |
+
value: 0.913
|
28 |
+
name: Recall
|
29 |
+
- type: f1
|
30 |
+
value: 0.9077802634849614
|
31 |
+
name: F1
|
32 |
+
- type: accuracy
|
33 |
+
value: 0.8330271015158475
|
34 |
+
name: Accuracy
|
35 |
---
|
36 |
+
|
37 |
+
# layoutlmv3-finetuned-funsd
|
38 |
+
|
39 |
+
This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on the pierreguillou/DocLayNet-large.
|
40 |
+
It achieves the following results on the evaluation set:
|
41 |
+
- Loss: 0.33888205885887146,
|
42 |
+
- Precision: 0.8478835766832817,
|
43 |
+
- Recall: 0.8934488524091807,
|
44 |
+
- F1: 0.8700700634847538,
|
45 |
+
- Accuracy: 0.9574140990541197
|
46 |
+
|
47 |
+
The script for training can be found here: https://github.com/huggingface/transformers/tree/main/examples/research_projects/layoutlmv3
|
48 |
+
|
49 |
+
## Model description
|
50 |
+
|
51 |
+
More information needed
|
52 |
+
|
53 |
+
## Intended uses & limitations
|
54 |
+
|
55 |
+
More information needed
|
56 |
+
|
57 |
+
## Training and evaluation data
|
58 |
+
|
59 |
+
More information needed
|
60 |
+
|
61 |
+
## Training procedure
|
62 |
+
|
63 |
+
### Training hyperparameters
|
64 |
+
|
65 |
+
The following hyperparameters were used during training:
|
66 |
+
- learning_rate: 1e-05
|
67 |
+
- train_batch_size: 2
|
68 |
+
- eval_batch_size: 2
|
69 |
+
- training_steps: 100000
|
70 |
+
|
71 |
+
### Framework versions
|
72 |
+
|
73 |
+
- Transformers 4.33.3
|
74 |
+
- Pytorch 1.11.0+cu115
|
75 |
+
- Datasets 2.14.5
|
76 |
+
- Tokenizers 0.13.3
|