balabis commited on
Commit
bdcb529
1 Parent(s): a32a54f

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +108 -0
README.md ADDED
@@ -0,0 +1,108 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-nc-sa-4.0
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - invoices
7
+ metrics:
8
+ - precision
9
+ - recall
10
+ - f1
11
+ - accuracy
12
+ model-index:
13
+ - name: layoutlmv3-finetuned-invoice
14
+ results:
15
+ - task:
16
+ name: Token Classification
17
+ type: token-classification
18
+ dataset:
19
+ name: invoices
20
+ type: invoices
21
+ config: sroie
22
+ split: train
23
+ args: sroie
24
+ metrics:
25
+ - name: Precision
26
+ type: precision
27
+ value: 0.975
28
+ - name: Recall
29
+ type: recall
30
+ value: 0.975
31
+ - name: F1
32
+ type: f1
33
+ value: 0.975
34
+ - name: Accuracy
35
+ type: accuracy
36
+ value: 0.975
37
+ ---
38
+
39
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
40
+ should probably proofread and complete it, then remove this comment. -->
41
+
42
+ # layoutlmv3-finetuned-invoice
43
+
44
+ This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on the invoices dataset.
45
+ It achieves the following results on the evaluation set:
46
+ - Loss: 0.2299
47
+ - Precision: 0.975
48
+ - Recall: 0.975
49
+ - F1: 0.975
50
+ - Accuracy: 0.975
51
+
52
+ ## Model description
53
+
54
+ More information needed
55
+
56
+ ## Intended uses & limitations
57
+
58
+ More information needed
59
+
60
+ ## Training and evaluation data
61
+
62
+ More information needed
63
+
64
+ ## Training procedure
65
+
66
+ ### Training hyperparameters
67
+
68
+ The following hyperparameters were used during training:
69
+ - learning_rate: 1e-05
70
+ - train_batch_size: 2
71
+ - eval_batch_size: 2
72
+ - seed: 42
73
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
74
+ - lr_scheduler_type: linear
75
+ - training_steps: 2000
76
+
77
+ ### Training results
78
+
79
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
80
+ |:-------------:|:------:|:----:|:---------------:|:---------:|:------:|:-----:|:--------:|
81
+ | No log | 14.29 | 100 | 0.1616 | 0.975 | 0.975 | 0.975 | 0.975 |
82
+ | No log | 28.57 | 200 | 0.1909 | 0.975 | 0.975 | 0.975 | 0.975 |
83
+ | No log | 42.86 | 300 | 0.2046 | 0.975 | 0.975 | 0.975 | 0.975 |
84
+ | No log | 57.14 | 400 | 0.2134 | 0.975 | 0.975 | 0.975 | 0.975 |
85
+ | 0.1239 | 71.43 | 500 | 0.2299 | 0.975 | 0.975 | 0.975 | 0.975 |
86
+ | 0.1239 | 85.71 | 600 | 0.2309 | 0.975 | 0.975 | 0.975 | 0.975 |
87
+ | 0.1239 | 100.0 | 700 | 0.2342 | 0.975 | 0.975 | 0.975 | 0.975 |
88
+ | 0.1239 | 114.29 | 800 | 0.2407 | 0.975 | 0.975 | 0.975 | 0.975 |
89
+ | 0.1239 | 128.57 | 900 | 0.2428 | 0.975 | 0.975 | 0.975 | 0.975 |
90
+ | 0.0007 | 142.86 | 1000 | 0.2449 | 0.975 | 0.975 | 0.975 | 0.975 |
91
+ | 0.0007 | 157.14 | 1100 | 0.2465 | 0.975 | 0.975 | 0.975 | 0.975 |
92
+ | 0.0007 | 171.43 | 1200 | 0.2488 | 0.975 | 0.975 | 0.975 | 0.975 |
93
+ | 0.0007 | 185.71 | 1300 | 0.2515 | 0.975 | 0.975 | 0.975 | 0.975 |
94
+ | 0.0007 | 200.0 | 1400 | 0.2525 | 0.975 | 0.975 | 0.975 | 0.975 |
95
+ | 0.0004 | 214.29 | 1500 | 0.2540 | 0.975 | 0.975 | 0.975 | 0.975 |
96
+ | 0.0004 | 228.57 | 1600 | 0.2557 | 0.975 | 0.975 | 0.975 | 0.975 |
97
+ | 0.0004 | 242.86 | 1700 | 0.2564 | 0.975 | 0.975 | 0.975 | 0.975 |
98
+ | 0.0004 | 257.14 | 1800 | 0.2570 | 0.975 | 0.975 | 0.975 | 0.975 |
99
+ | 0.0004 | 271.43 | 1900 | 0.2573 | 0.975 | 0.975 | 0.975 | 0.975 |
100
+ | 0.0003 | 285.71 | 2000 | 0.2574 | 0.975 | 0.975 | 0.975 | 0.975 |
101
+
102
+
103
+ ### Framework versions
104
+
105
+ - Transformers 4.23.0.dev0
106
+ - Pytorch 1.12.1+cu113
107
+ - Datasets 2.4.0
108
+ - Tokenizers 0.12.1