Theivaprakasham commited on
Commit
5100120
1 Parent(s): 8723476

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +125 -0
README.md ADDED
@@ -0,0 +1,125 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - generated_from_trainer
4
+ datasets:
5
+ - wild_receipt
6
+ metrics:
7
+ - precision
8
+ - recall
9
+ - f1
10
+ - accuracy
11
+ model-index:
12
+ - name: layoutlmv3-finetuned-wildreceipt
13
+ results:
14
+ - task:
15
+ name: Token Classification
16
+ type: token-classification
17
+ dataset:
18
+ name: wild_receipt
19
+ type: wild_receipt
20
+ args: WildReceipt
21
+ metrics:
22
+ - name: Precision
23
+ type: precision
24
+ value: 0.877212237618329
25
+ - name: Recall
26
+ type: recall
27
+ value: 0.8798678959680749
28
+ - name: F1
29
+ type: f1
30
+ value: 0.8785380599065679
31
+ - name: Accuracy
32
+ type: accuracy
33
+ value: 0.9249204782274871
34
+ ---
35
+
36
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
37
+ should probably proofread and complete it, then remove this comment. -->
38
+
39
+ # layoutlmv3-finetuned-wildreceipt
40
+
41
+ This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on the wild_receipt dataset.
42
+ It achieves the following results on the evaluation set:
43
+ - Loss: 0.3108
44
+ - Precision: 0.8772
45
+ - Recall: 0.8799
46
+ - F1: 0.8785
47
+ - Accuracy: 0.9249
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: 4
68
+ - eval_batch_size: 4
69
+ - seed: 42
70
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
71
+ - lr_scheduler_type: linear
72
+ - training_steps: 4000
73
+
74
+ ### Training results
75
+
76
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
77
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
78
+ | No log | 0.32 | 100 | 1.3143 | 0.6709 | 0.2679 | 0.3829 | 0.6700 |
79
+ | No log | 0.63 | 200 | 0.8814 | 0.6478 | 0.5195 | 0.5766 | 0.7786 |
80
+ | No log | 0.95 | 300 | 0.6568 | 0.7205 | 0.6491 | 0.6829 | 0.8303 |
81
+ | No log | 1.26 | 400 | 0.5618 | 0.7544 | 0.7072 | 0.7300 | 0.8519 |
82
+ | 1.0284 | 1.58 | 500 | 0.5003 | 0.7802 | 0.7566 | 0.7682 | 0.8687 |
83
+ | 1.0284 | 1.89 | 600 | 0.4454 | 0.7941 | 0.7679 | 0.7807 | 0.8748 |
84
+ | 1.0284 | 2.21 | 700 | 0.4314 | 0.8142 | 0.7928 | 0.8033 | 0.8852 |
85
+ | 1.0284 | 2.52 | 800 | 0.3870 | 0.8172 | 0.8200 | 0.8186 | 0.8953 |
86
+ | 1.0284 | 2.84 | 900 | 0.3629 | 0.8288 | 0.8369 | 0.8329 | 0.9025 |
87
+ | 0.4167 | 3.15 | 1000 | 0.3537 | 0.8540 | 0.8200 | 0.8366 | 0.9052 |
88
+ | 0.4167 | 3.47 | 1100 | 0.3383 | 0.8438 | 0.8285 | 0.8361 | 0.9063 |
89
+ | 0.4167 | 3.79 | 1200 | 0.3403 | 0.8297 | 0.8493 | 0.8394 | 0.9062 |
90
+ | 0.4167 | 4.1 | 1300 | 0.3271 | 0.8428 | 0.8545 | 0.8487 | 0.9110 |
91
+ | 0.4167 | 4.42 | 1400 | 0.3182 | 0.8491 | 0.8518 | 0.8504 | 0.9131 |
92
+ | 0.2766 | 4.73 | 1500 | 0.3111 | 0.8491 | 0.8539 | 0.8515 | 0.9129 |
93
+ | 0.2766 | 5.05 | 1600 | 0.3177 | 0.8397 | 0.8620 | 0.8507 | 0.9124 |
94
+ | 0.2766 | 5.36 | 1700 | 0.3091 | 0.8676 | 0.8548 | 0.8612 | 0.9191 |
95
+ | 0.2766 | 5.68 | 1800 | 0.3080 | 0.8508 | 0.8645 | 0.8576 | 0.9162 |
96
+ | 0.2766 | 5.99 | 1900 | 0.3059 | 0.8492 | 0.8662 | 0.8576 | 0.9163 |
97
+ | 0.2114 | 6.31 | 2000 | 0.3184 | 0.8536 | 0.8657 | 0.8596 | 0.9147 |
98
+ | 0.2114 | 6.62 | 2100 | 0.3161 | 0.8583 | 0.8713 | 0.8648 | 0.9184 |
99
+ | 0.2114 | 6.94 | 2200 | 0.3055 | 0.8707 | 0.8682 | 0.8694 | 0.9220 |
100
+ | 0.2114 | 7.26 | 2300 | 0.3004 | 0.8689 | 0.8745 | 0.8717 | 0.9219 |
101
+ | 0.2114 | 7.57 | 2400 | 0.3111 | 0.8701 | 0.8720 | 0.8711 | 0.9211 |
102
+ | 0.174 | 7.89 | 2500 | 0.3130 | 0.8599 | 0.8741 | 0.8669 | 0.9198 |
103
+ | 0.174 | 8.2 | 2600 | 0.3034 | 0.8661 | 0.8748 | 0.8704 | 0.9219 |
104
+ | 0.174 | 8.52 | 2700 | 0.3005 | 0.8799 | 0.8673 | 0.8736 | 0.9225 |
105
+ | 0.174 | 8.83 | 2800 | 0.3043 | 0.8687 | 0.8804 | 0.8745 | 0.9240 |
106
+ | 0.174 | 9.15 | 2900 | 0.3121 | 0.8776 | 0.8704 | 0.8740 | 0.9242 |
107
+ | 0.1412 | 9.46 | 3000 | 0.3131 | 0.8631 | 0.8755 | 0.8692 | 0.9204 |
108
+ | 0.1412 | 9.78 | 3100 | 0.3067 | 0.8715 | 0.8773 | 0.8744 | 0.9233 |
109
+ | 0.1412 | 10.09 | 3200 | 0.3021 | 0.8751 | 0.8812 | 0.8782 | 0.9248 |
110
+ | 0.1412 | 10.41 | 3300 | 0.3092 | 0.8651 | 0.8808 | 0.8729 | 0.9228 |
111
+ | 0.1412 | 10.73 | 3400 | 0.3084 | 0.8776 | 0.8749 | 0.8762 | 0.9237 |
112
+ | 0.1254 | 11.04 | 3500 | 0.3156 | 0.8738 | 0.8785 | 0.8761 | 0.9237 |
113
+ | 0.1254 | 11.36 | 3600 | 0.3131 | 0.8723 | 0.8818 | 0.8770 | 0.9244 |
114
+ | 0.1254 | 11.67 | 3700 | 0.3108 | 0.8778 | 0.8781 | 0.8780 | 0.9250 |
115
+ | 0.1254 | 11.99 | 3800 | 0.3097 | 0.8778 | 0.8771 | 0.8775 | 0.9239 |
116
+ | 0.1254 | 12.3 | 3900 | 0.3115 | 0.8785 | 0.8801 | 0.8793 | 0.9251 |
117
+ | 0.111 | 12.62 | 4000 | 0.3108 | 0.8772 | 0.8799 | 0.8785 | 0.9249 |
118
+
119
+
120
+ ### Framework versions
121
+
122
+ - Transformers 4.20.0.dev0
123
+ - Pytorch 1.11.0+cu113
124
+ - Datasets 2.2.2
125
+ - Tokenizers 0.12.1