Hugging Face's logo

Saed2023
/
layoutlmv3-finetuned-Algo_427Images

Token Classification
Transformers
PyTorch
TensorBoard
layoutlmv3
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
1
Saed2023 commited on
Commit
c88e097
·
1 Parent(s): f82e84e

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +16 -56
README.md CHANGED
@@ -19,11 +19,11 @@ should probably proofread and complete it, then remove this comment. -->
19
 
20
  This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.0019
23
- - Precision: 0.9891
24
- - Recall: 0.9909
25
- - F1: 0.9900
26
- - Accuracy: 0.9997
27
 
28
  ## Model description
29
 
@@ -48,62 +48,22 @@ The following hyperparameters were used during training:
48
  - seed: 42
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: linear
51
- - training_steps: 500
52
 
53
  ### Training results
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
56
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
57
- | No log | 0.12 | 10 | 0.1986 | 0.0 | 0.0 | 0.0 | 0.9661 |
58
- | No log | 0.25 | 20 | 0.1131 | 0.0 | 0.0 | 0.0 | 0.9661 |
59
- | No log | 0.38 | 30 | 0.0757 | 0.1848 | 0.1109 | 0.1386 | 0.9722 |
60
- | No log | 0.5 | 40 | 0.0600 | 0.4032 | 0.0909 | 0.1484 | 0.9784 |
61
- | No log | 0.62 | 50 | 0.0481 | 0.6446 | 0.3891 | 0.4853 | 0.9869 |
62
- | No log | 0.75 | 60 | 0.0384 | 0.8022 | 0.6709 | 0.7307 | 0.9924 |
63
- | No log | 0.88 | 70 | 0.0276 | 0.8347 | 0.7527 | 0.7916 | 0.9949 |
64
- | No log | 1.0 | 80 | 0.0194 | 0.8333 | 0.7545 | 0.7920 | 0.9949 |
65
- | No log | 1.12 | 90 | 0.0137 | 0.9118 | 0.8836 | 0.8975 | 0.9973 |
66
- | No log | 1.25 | 100 | 0.0105 | 0.95 | 0.9327 | 0.9413 | 0.9983 |
67
- | No log | 1.38 | 110 | 0.0080 | 0.9557 | 0.9418 | 0.9487 | 0.9986 |
68
- | No log | 1.5 | 120 | 0.0068 | 0.9650 | 0.9527 | 0.9588 | 0.9989 |
69
- | No log | 1.62 | 130 | 0.0055 | 0.9741 | 0.9564 | 0.9651 | 0.9990 |
70
- | No log | 1.75 | 140 | 0.0048 | 0.9745 | 0.9709 | 0.9727 | 0.9993 |
71
- | No log | 1.88 | 150 | 0.0043 | 0.9781 | 0.9727 | 0.9754 | 0.9993 |
72
- | No log | 2.0 | 160 | 0.0037 | 0.9817 | 0.9727 | 0.9772 | 0.9993 |
73
- | No log | 2.12 | 170 | 0.0034 | 0.9835 | 0.9782 | 0.9809 | 0.9994 |
74
- | No log | 2.25 | 180 | 0.0037 | 0.9762 | 0.9691 | 0.9726 | 0.9993 |
75
- | No log | 2.38 | 190 | 0.0030 | 0.9855 | 0.9855 | 0.9855 | 0.9996 |
76
- | No log | 2.5 | 200 | 0.0030 | 0.9854 | 0.9836 | 0.9845 | 0.9995 |
77
- | No log | 2.62 | 210 | 0.0029 | 0.9855 | 0.9855 | 0.9855 | 0.9996 |
78
- | No log | 2.75 | 220 | 0.0027 | 0.9836 | 0.9818 | 0.9827 | 0.9994 |
79
- | No log | 2.88 | 230 | 0.0026 | 0.9854 | 0.9818 | 0.9836 | 0.9994 |
80
- | No log | 3.0 | 240 | 0.0025 | 0.9873 | 0.9891 | 0.9882 | 0.9996 |
81
- | No log | 3.12 | 250 | 0.0025 | 0.9873 | 0.9891 | 0.9882 | 0.9996 |
82
- | No log | 3.25 | 260 | 0.0023 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
83
- | No log | 3.38 | 270 | 0.0024 | 0.9891 | 0.9891 | 0.9891 | 0.9996 |
84
- | No log | 3.5 | 280 | 0.0023 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
85
- | No log | 3.62 | 290 | 0.0023 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
86
- | No log | 3.75 | 300 | 0.0022 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
87
- | No log | 3.88 | 310 | 0.0021 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
88
- | No log | 4.0 | 320 | 0.0021 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
89
- | No log | 4.12 | 330 | 0.0020 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
90
- | No log | 4.25 | 340 | 0.0020 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
91
- | No log | 4.38 | 350 | 0.0020 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
92
- | No log | 4.5 | 360 | 0.0020 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
93
- | No log | 4.62 | 370 | 0.0020 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
94
- | No log | 4.75 | 380 | 0.0019 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
95
- | No log | 4.88 | 390 | 0.0019 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
96
- | No log | 5.0 | 400 | 0.0019 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
97
- | No log | 5.12 | 410 | 0.0019 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
98
- | No log | 5.25 | 420 | 0.0019 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
99
- | No log | 5.38 | 430 | 0.0019 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
100
- | No log | 5.5 | 440 | 0.0019 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
101
- | No log | 5.62 | 450 | 0.0019 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
102
- | No log | 5.75 | 460 | 0.0019 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
103
- | No log | 5.88 | 470 | 0.0019 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
104
- | No log | 6.0 | 480 | 0.0019 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
105
- | No log | 6.12 | 490 | 0.0019 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
106
- | 0.0346 | 6.25 | 500 | 0.0019 | 0.9891 | 0.9909 | 0.9900 | 0.9997 |
107
 
108
 
109
  ### Framework versions
 
19
 
20
  This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.0013
23
+ - Precision: 0.9937
24
+ - Recall: 0.9964
25
+ - F1: 0.9950
26
+ - Accuracy: 0.9999
27
 
28
  ## Model description
29
 
 
48
  - seed: 42
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: linear
51
+ - training_steps: 1000
52
 
53
  ### Training results
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
56
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
57
+ | No log | 1.25 | 100 | 0.0082 | 0.9505 | 0.9367 | 0.9435 | 0.9983 |
58
+ | No log | 2.5 | 200 | 0.0024 | 0.9883 | 0.9901 | 0.9892 | 0.9997 |
59
+ | No log | 3.75 | 300 | 0.0020 | 0.9883 | 0.9919 | 0.9901 | 0.9997 |
60
+ | No log | 5.0 | 400 | 0.0016 | 0.9910 | 0.9928 | 0.9919 | 0.9998 |
61
+ | 0.0301 | 6.25 | 500 | 0.0015 | 0.9910 | 0.9928 | 0.9919 | 0.9998 |
62
+ | 0.0301 | 7.5 | 600 | 0.0014 | 0.9928 | 0.9946 | 0.9937 | 0.9998 |
63
+ | 0.0301 | 8.75 | 700 | 0.0013 | 0.9928 | 0.9946 | 0.9937 | 0.9998 |
64
+ | 0.0301 | 10.0 | 800 | 0.0013 | 0.9937 | 0.9964 | 0.9950 | 0.9999 |
65
+ | 0.0301 | 11.25 | 900 | 0.0013 | 0.9928 | 0.9946 | 0.9937 | 0.9998 |
66
+ | 0.002 | 12.5 | 1000 | 0.0013 | 0.9937 | 0.9964 | 0.9950 | 0.9999 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
67
 
68
 
69
  ### Framework versions