Hugging Face's logo

AliShaker
/
layoutlmv3-finetuned-wildreceipt

Token Classification
Transformers
PyTorch
TensorBoard
layoutlmv3
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
2
AliShaker commited on
Commit
31be7d0
•
1 Parent(s): db2885a

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +128 -0
README.md ADDED
@@ -0,0 +1,128 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-nc-sa-4.0
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - wildreceipt
7
+ metrics:
8
+ - precision
9
+ - recall
10
+ - f1
11
+ - accuracy
12
+ model-index:
13
+ - name: layoutlmv3-finetuned-wildreceipt
14
+ results:
15
+ - task:
16
+ name: Token Classification
17
+ type: token-classification
18
+ dataset:
19
+ name: wildreceipt
20
+ type: wildreceipt
21
+ config: WildReceipt
22
+ split: train
23
+ args: WildReceipt
24
+ metrics:
25
+ - name: Precision
26
+ type: precision
27
+ value: 0.877962408063198
28
+ - name: Recall
29
+ type: recall
30
+ value: 0.8870235310306867
31
+ - name: F1
32
+ type: f1
33
+ value: 0.8824697104524608
34
+ - name: Accuracy
35
+ type: accuracy
36
+ value: 0.9265109136777449
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-wildreceipt
43
+
44
+ This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on the wildreceipt dataset.
45
+ It achieves the following results on the evaluation set:
46
+ - Loss: 0.3129
47
+ - Precision: 0.8780
48
+ - Recall: 0.8870
49
+ - F1: 0.8825
50
+ - Accuracy: 0.9265
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: 4
71
+ - eval_batch_size: 4
72
+ - seed: 42
73
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
74
+ - lr_scheduler_type: linear
75
+ - training_steps: 4000
76
+
77
+ ### Training results
78
+
79
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
80
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
81
+ | No log | 0.32 | 100 | 1.2240 | 0.6077 | 0.3766 | 0.4650 | 0.7011 |
82
+ | No log | 0.63 | 200 | 0.8417 | 0.6440 | 0.5089 | 0.5685 | 0.7743 |
83
+ | No log | 0.95 | 300 | 0.6466 | 0.7243 | 0.6583 | 0.6897 | 0.8311 |
84
+ | No log | 1.26 | 400 | 0.5516 | 0.7533 | 0.7158 | 0.7341 | 0.8537 |
85
+ | 0.9961 | 1.58 | 500 | 0.4845 | 0.7835 | 0.7557 | 0.7693 | 0.8699 |
86
+ | 0.9961 | 1.89 | 600 | 0.4506 | 0.7809 | 0.7930 | 0.7869 | 0.8770 |
87
+ | 0.9961 | 2.21 | 700 | 0.4230 | 0.8101 | 0.8107 | 0.8104 | 0.8886 |
88
+ | 0.9961 | 2.52 | 800 | 0.3797 | 0.8211 | 0.8296 | 0.8253 | 0.8983 |
89
+ | 0.9961 | 2.84 | 900 | 0.3576 | 0.8289 | 0.8411 | 0.8349 | 0.9016 |
90
+ | 0.4076 | 3.15 | 1000 | 0.3430 | 0.8394 | 0.8371 | 0.8382 | 0.9055 |
91
+ | 0.4076 | 3.47 | 1100 | 0.3354 | 0.8531 | 0.8405 | 0.8467 | 0.9071 |
92
+ | 0.4076 | 3.79 | 1200 | 0.3331 | 0.8371 | 0.8504 | 0.8437 | 0.9076 |
93
+ | 0.4076 | 4.1 | 1300 | 0.3184 | 0.8445 | 0.8609 | 0.8526 | 0.9118 |
94
+ | 0.4076 | 4.42 | 1400 | 0.3087 | 0.8617 | 0.8580 | 0.8598 | 0.9150 |
95
+ | 0.2673 | 4.73 | 1500 | 0.3013 | 0.8613 | 0.8657 | 0.8635 | 0.9177 |
96
+ | 0.2673 | 5.05 | 1600 | 0.2971 | 0.8630 | 0.8689 | 0.8659 | 0.9181 |
97
+ | 0.2673 | 5.36 | 1700 | 0.3075 | 0.8675 | 0.8639 | 0.8657 | 0.9177 |
98
+ | 0.2673 | 5.68 | 1800 | 0.2989 | 0.8551 | 0.8764 | 0.8656 | 0.9193 |
99
+ | 0.2673 | 5.99 | 1900 | 0.3011 | 0.8572 | 0.8762 | 0.8666 | 0.9194 |
100
+ | 0.2026 | 6.31 | 2000 | 0.3107 | 0.8595 | 0.8722 | 0.8658 | 0.9181 |
101
+ | 0.2026 | 6.62 | 2100 | 0.3050 | 0.8678 | 0.8800 | 0.8739 | 0.9220 |
102
+ | 0.2026 | 6.94 | 2200 | 0.2971 | 0.8722 | 0.8789 | 0.8755 | 0.9237 |
103
+ | 0.2026 | 7.26 | 2300 | 0.3057 | 0.8666 | 0.8785 | 0.8725 | 0.9209 |
104
+ | 0.2026 | 7.57 | 2400 | 0.3172 | 0.8593 | 0.8773 | 0.8682 | 0.9184 |
105
+ | 0.1647 | 7.89 | 2500 | 0.3018 | 0.8695 | 0.8823 | 0.8759 | 0.9228 |
106
+ | 0.1647 | 8.2 | 2600 | 0.3001 | 0.8760 | 0.8795 | 0.8777 | 0.9256 |
107
+ | 0.1647 | 8.52 | 2700 | 0.3068 | 0.8758 | 0.8745 | 0.8752 | 0.9235 |
108
+ | 0.1647 | 8.83 | 2800 | 0.3007 | 0.8779 | 0.8779 | 0.8779 | 0.9248 |
109
+ | 0.1647 | 9.15 | 2900 | 0.3063 | 0.8740 | 0.8763 | 0.8751 | 0.9228 |
110
+ | 0.1342 | 9.46 | 3000 | 0.3096 | 0.8675 | 0.8834 | 0.8754 | 0.9235 |
111
+ | 0.1342 | 9.78 | 3100 | 0.3052 | 0.8736 | 0.8848 | 0.8792 | 0.9249 |
112
+ | 0.1342 | 10.09 | 3200 | 0.3120 | 0.8727 | 0.8885 | 0.8805 | 0.9252 |
113
+ | 0.1342 | 10.41 | 3300 | 0.3146 | 0.8718 | 0.8843 | 0.8780 | 0.9243 |
114
+ | 0.1342 | 10.73 | 3400 | 0.3124 | 0.8720 | 0.8880 | 0.8799 | 0.9253 |
115
+ | 0.117 | 11.04 | 3500 | 0.3088 | 0.8761 | 0.8817 | 0.8789 | 0.9252 |
116
+ | 0.117 | 11.36 | 3600 | 0.3082 | 0.8782 | 0.8834 | 0.8808 | 0.9257 |
117
+ | 0.117 | 11.67 | 3700 | 0.3129 | 0.8767 | 0.8847 | 0.8807 | 0.9256 |
118
+ | 0.117 | 11.99 | 3800 | 0.3116 | 0.8792 | 0.8847 | 0.8820 | 0.9265 |
119
+ | 0.117 | 12.3 | 3900 | 0.3142 | 0.8768 | 0.8874 | 0.8821 | 0.9261 |
120
+ | 0.1022 | 12.62 | 4000 | 0.3129 | 0.8780 | 0.8870 | 0.8825 | 0.9265 |
121
+
122
+
123
+ ### Framework versions
124
+
125
+ - Transformers 4.22.0.dev0
126
+ - Pytorch 1.12.1+cu113
127
+ - Datasets 2.4.0
128
+ - Tokenizers 0.12.1