File size: 19,121 Bytes
1b9a513
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3ef071e
 
 
 
 
 
 
1b9a513
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3ef071e
1b9a513
 
 
 
3ef071e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1b9a513
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
---
license: mit
base_model: microsoft/layoutlm-base-uncased
tags:
- generated_from_trainer
model-index:
- name: layoutlm-custom_no_text
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# layoutlm-custom_no_text

This model is a fine-tuned version of [microsoft/layoutlm-base-uncased](https://huggingface.co/microsoft/layoutlm-base-uncased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5142
- Noise: {'precision': 0.6764705882352942, 'recall': 0.6695205479452054, 'f1': 0.6729776247848538, 'number': 584}
- Signal: {'precision': 0.629757785467128, 'recall': 0.6232876712328768, 'f1': 0.6265060240963856, 'number': 584}
- Overall Precision: 0.6531
- Overall Recall: 0.6464
- Overall F1: 0.6497
- Overall Accuracy: 0.9156

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Noise                                                                                                      | Signal                                                                                                     | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:----------------------------------------------------------------------------------------------------------:|:----------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
| 0.5712        | 1.0   | 18   | 0.4356          | {'precision': 0.3712374581939799, 'recall': 0.3801369863013699, 'f1': 0.3756345177664975, 'number': 584}   | {'precision': 0.3294314381270903, 'recall': 0.3373287671232877, 'f1': 0.3333333333333333, 'number': 584}   | 0.3503            | 0.3587         | 0.3545     | 0.8008           |
| 0.4233        | 2.0   | 36   | 0.3745          | {'precision': 0.4048964218455744, 'recall': 0.3681506849315068, 'f1': 0.38565022421524664, 'number': 584}  | {'precision': 0.3483992467043315, 'recall': 0.3167808219178082, 'f1': 0.33183856502242154, 'number': 584}  | 0.3766            | 0.3425         | 0.3587     | 0.8287           |
| 0.3817        | 3.0   | 54   | 0.3632          | {'precision': 0.45740740740740743, 'recall': 0.4229452054794521, 'f1': 0.43950177935943063, 'number': 584} | {'precision': 0.3851851851851852, 'recall': 0.3561643835616438, 'f1': 0.3701067615658363, 'number': 584}   | 0.4213            | 0.3896         | 0.4048     | 0.8413           |
| 0.3472        | 4.0   | 72   | 0.3133          | {'precision': 0.5143953934740882, 'recall': 0.4589041095890411, 'f1': 0.4850678733031674, 'number': 584}   | {'precision': 0.43378119001919385, 'recall': 0.386986301369863, 'f1': 0.40904977375565615, 'number': 584}  | 0.4741            | 0.4229         | 0.4471     | 0.8550           |
| 0.3132        | 5.0   | 90   | 0.3254          | {'precision': 0.5112016293279023, 'recall': 0.4297945205479452, 'f1': 0.4669767441860465, 'number': 584}   | {'precision': 0.4460285132382892, 'recall': 0.375, 'f1': 0.4074418604651162, 'number': 584}                | 0.4786            | 0.4024         | 0.4372     | 0.8525           |
| 0.282         | 6.0   | 108  | 0.3033          | {'precision': 0.5387453874538746, 'recall': 0.5, 'f1': 0.5186500888099467, 'number': 584}                  | {'precision': 0.46863468634686345, 'recall': 0.4349315068493151, 'f1': 0.45115452930728245, 'number': 584} | 0.5037            | 0.4675         | 0.4849     | 0.8656           |
| 0.2486        | 7.0   | 126  | 0.2827          | {'precision': 0.5498220640569395, 'recall': 0.5291095890410958, 'f1': 0.5392670157068062, 'number': 584}   | {'precision': 0.5071174377224199, 'recall': 0.488013698630137, 'f1': 0.49738219895287955, 'number': 584}   | 0.5285            | 0.5086         | 0.5183     | 0.8773           |
| 0.2276        | 8.0   | 144  | 0.2798          | {'precision': 0.5597826086956522, 'recall': 0.5291095890410958, 'f1': 0.5440140845070423, 'number': 584}   | {'precision': 0.5235507246376812, 'recall': 0.4948630136986301, 'f1': 0.5088028169014084, 'number': 584}   | 0.5417            | 0.5120         | 0.5264     | 0.8793           |
| 0.197         | 9.0   | 162  | 0.2778          | {'precision': 0.5948905109489051, 'recall': 0.5582191780821918, 'f1': 0.5759717314487632, 'number': 584}   | {'precision': 0.5602189781021898, 'recall': 0.5256849315068494, 'f1': 0.5424028268551236, 'number': 584}   | 0.5776            | 0.5420         | 0.5592     | 0.8891           |
| 0.1812        | 10.0  | 180  | 0.2932          | {'precision': 0.5907407407407408, 'recall': 0.5462328767123288, 'f1': 0.5676156583629893, 'number': 584}   | {'precision': 0.5666666666666667, 'recall': 0.523972602739726, 'f1': 0.5444839857651246, 'number': 584}    | 0.5787            | 0.5351         | 0.5560     | 0.8888           |
| 0.1611        | 11.0  | 198  | 0.2785          | {'precision': 0.6156648451730419, 'recall': 0.5787671232876712, 'f1': 0.5966460723742276, 'number': 584}   | {'precision': 0.5719489981785064, 'recall': 0.5376712328767124, 'f1': 0.5542806707855252, 'number': 584}   | 0.5938            | 0.5582         | 0.5755     | 0.8991           |
| 0.1441        | 12.0  | 216  | 0.2738          | {'precision': 0.6263537906137184, 'recall': 0.5941780821917808, 'f1': 0.6098418277680141, 'number': 584}   | {'precision': 0.5776173285198556, 'recall': 0.547945205479452, 'f1': 0.562390158172232, 'number': 584}     | 0.6020            | 0.5711         | 0.5861     | 0.9016           |
| 0.1294        | 13.0  | 234  | 0.3072          | {'precision': 0.6201413427561837, 'recall': 0.601027397260274, 'f1': 0.6104347826086957, 'number': 584}    | {'precision': 0.5795053003533569, 'recall': 0.5616438356164384, 'f1': 0.5704347826086956, 'number': 584}   | 0.5998            | 0.5813         | 0.5904     | 0.8989           |
| 0.1218        | 14.0  | 252  | 0.2963          | {'precision': 0.629695885509839, 'recall': 0.6027397260273972, 'f1': 0.6159230096237971, 'number': 584}    | {'precision': 0.5849731663685152, 'recall': 0.559931506849315, 'f1': 0.5721784776902886, 'number': 584}    | 0.6073            | 0.5813         | 0.5941     | 0.9030           |
| 0.1032        | 15.0  | 270  | 0.3365          | {'precision': 0.6106194690265486, 'recall': 0.5907534246575342, 'f1': 0.6005221932114881, 'number': 584}   | {'precision': 0.5681415929203539, 'recall': 0.5496575342465754, 'f1': 0.5587467362924282, 'number': 584}   | 0.5894            | 0.5702         | 0.5796     | 0.8991           |
| 0.0981        | 16.0  | 288  | 0.3342          | {'precision': 0.631858407079646, 'recall': 0.6113013698630136, 'f1': 0.6214099216710183, 'number': 584}    | {'precision': 0.5893805309734513, 'recall': 0.5702054794520548, 'f1': 0.5796344647519582, 'number': 584}   | 0.6106            | 0.5908         | 0.6005     | 0.9039           |
| 0.0844        | 17.0  | 306  | 0.3543          | {'precision': 0.6502636203866432, 'recall': 0.6335616438356164, 'f1': 0.6418039895923676, 'number': 584}   | {'precision': 0.5957820738137083, 'recall': 0.5804794520547946, 'f1': 0.5880312228967911, 'number': 584}   | 0.6230            | 0.6070         | 0.6149     | 0.9050           |
| 0.0763        | 18.0  | 324  | 0.3559          | {'precision': 0.6392294220665499, 'recall': 0.625, 'f1': 0.632034632034632, 'number': 584}                 | {'precision': 0.5989492119089317, 'recall': 0.5856164383561644, 'f1': 0.5922077922077922, 'number': 584}   | 0.6191            | 0.6053         | 0.6121     | 0.9075           |
| 0.0682        | 19.0  | 342  | 0.3599          | {'precision': 0.6666666666666666, 'recall': 0.6335616438356164, 'f1': 0.6496927129060578, 'number': 584}   | {'precision': 0.618018018018018, 'recall': 0.5873287671232876, 'f1': 0.6022827041264267, 'number': 584}    | 0.6423            | 0.6104         | 0.6260     | 0.9086           |
| 0.0685        | 20.0  | 360  | 0.3574          | {'precision': 0.670863309352518, 'recall': 0.6386986301369864, 'f1': 0.6543859649122807, 'number': 584}    | {'precision': 0.6151079136690647, 'recall': 0.5856164383561644, 'f1': 0.6, 'number': 584}                  | 0.6430            | 0.6122         | 0.6272     | 0.9114           |
| 0.0591        | 21.0  | 378  | 0.3742          | {'precision': 0.6573426573426573, 'recall': 0.6438356164383562, 'f1': 0.6505190311418684, 'number': 584}   | {'precision': 0.6171328671328671, 'recall': 0.6044520547945206, 'f1': 0.6107266435986158, 'number': 584}   | 0.6372            | 0.6241         | 0.6306     | 0.9100           |
| 0.0521        | 22.0  | 396  | 0.4063          | {'precision': 0.6566901408450704, 'recall': 0.6386986301369864, 'f1': 0.6475694444444444, 'number': 584}   | {'precision': 0.6161971830985915, 'recall': 0.5993150684931506, 'f1': 0.607638888888889, 'number': 584}    | 0.6364            | 0.6190         | 0.6276     | 0.9095           |
| 0.0492        | 23.0  | 414  | 0.3971          | {'precision': 0.649737302977233, 'recall': 0.6352739726027398, 'f1': 0.6424242424242426, 'number': 584}    | {'precision': 0.5971978984238179, 'recall': 0.583904109589041, 'f1': 0.5904761904761905, 'number': 584}    | 0.6235            | 0.6096         | 0.6165     | 0.9086           |
| 0.045         | 24.0  | 432  | 0.4198          | {'precision': 0.6448275862068965, 'recall': 0.6404109589041096, 'f1': 0.6426116838487972, 'number': 584}   | {'precision': 0.5948275862068966, 'recall': 0.5907534246575342, 'f1': 0.5927835051546393, 'number': 584}   | 0.6198            | 0.6156         | 0.6177     | 0.9061           |
| 0.0391        | 25.0  | 450  | 0.4477          | {'precision': 0.643979057591623, 'recall': 0.6318493150684932, 'f1': 0.6378565254969749, 'number': 584}    | {'precision': 0.5986038394415357, 'recall': 0.5873287671232876, 'f1': 0.5929127052722557, 'number': 584}   | 0.6213            | 0.6096         | 0.6154     | 0.9061           |
| 0.0411        | 26.0  | 468  | 0.4080          | {'precision': 0.6400679117147708, 'recall': 0.6455479452054794, 'f1': 0.6427962489343563, 'number': 584}   | {'precision': 0.597623089983022, 'recall': 0.6027397260273972, 'f1': 0.6001705029838021, 'number': 584}    | 0.6188            | 0.6241         | 0.6215     | 0.9084           |
| 0.0369        | 27.0  | 486  | 0.4339          | {'precision': 0.6614035087719298, 'recall': 0.6455479452054794, 'f1': 0.6533795493934141, 'number': 584}   | {'precision': 0.6105263157894737, 'recall': 0.5958904109589042, 'f1': 0.6031195840554593, 'number': 584}   | 0.6360            | 0.6207         | 0.6282     | 0.9103           |
| 0.0315        | 28.0  | 504  | 0.4303          | {'precision': 0.6637931034482759, 'recall': 0.6592465753424658, 'f1': 0.6615120274914089, 'number': 584}   | {'precision': 0.6137931034482759, 'recall': 0.6095890410958904, 'f1': 0.6116838487972508, 'number': 584}   | 0.6388            | 0.6344         | 0.6366     | 0.9117           |
| 0.0332        | 29.0  | 522  | 0.4253          | {'precision': 0.6643717728055077, 'recall': 0.660958904109589, 'f1': 0.6626609442060085, 'number': 584}    | {'precision': 0.6179001721170396, 'recall': 0.6147260273972602, 'f1': 0.6163090128755364, 'number': 584}   | 0.6411            | 0.6378         | 0.6395     | 0.9134           |
| 0.0272        | 30.0  | 540  | 0.4594          | {'precision': 0.6495726495726496, 'recall': 0.6506849315068494, 'f1': 0.6501283147989735, 'number': 584}   | {'precision': 0.5931623931623932, 'recall': 0.5941780821917808, 'f1': 0.5936698032506416, 'number': 584}   | 0.6214            | 0.6224         | 0.6219     | 0.9078           |
| 0.027         | 31.0  | 558  | 0.4680          | {'precision': 0.6621160409556314, 'recall': 0.6643835616438356, 'f1': 0.6632478632478632, 'number': 584}   | {'precision': 0.6143344709897611, 'recall': 0.6164383561643836, 'f1': 0.6153846153846154, 'number': 584}   | 0.6382            | 0.6404         | 0.6393     | 0.9111           |
| 0.0295        | 32.0  | 576  | 0.4367          | {'precision': 0.6719022687609075, 'recall': 0.6592465753424658, 'f1': 0.6655142610198791, 'number': 584}   | {'precision': 0.612565445026178, 'recall': 0.601027397260274, 'f1': 0.6067415730337079, 'number': 584}     | 0.6422            | 0.6301         | 0.6361     | 0.9120           |
| 0.0216        | 33.0  | 594  | 0.4674          | {'precision': 0.681260945709282, 'recall': 0.666095890410959, 'f1': 0.6735930735930735, 'number': 584}     | {'precision': 0.6357267950963222, 'recall': 0.6215753424657534, 'f1': 0.6285714285714286, 'number': 584}   | 0.6585            | 0.6438         | 0.6511     | 0.9139           |
| 0.0212        | 34.0  | 612  | 0.4702          | {'precision': 0.6666666666666666, 'recall': 0.6643835616438356, 'f1': 0.6655231560891938, 'number': 584}   | {'precision': 0.6202749140893471, 'recall': 0.6181506849315068, 'f1': 0.6192109777015438, 'number': 584}   | 0.6435            | 0.6413         | 0.6424     | 0.9103           |
| 0.0227        | 35.0  | 630  | 0.4637          | {'precision': 0.657672849915683, 'recall': 0.6678082191780822, 'f1': 0.6627017841971112, 'number': 584}    | {'precision': 0.6155143338954469, 'recall': 0.625, 'f1': 0.6202209005947323, 'number': 584}                | 0.6366            | 0.6464         | 0.6415     | 0.9109           |
| 0.0196        | 36.0  | 648  | 0.4639          | {'precision': 0.6660899653979239, 'recall': 0.6592465753424658, 'f1': 0.6626506024096386, 'number': 584}   | {'precision': 0.6141868512110726, 'recall': 0.6078767123287672, 'f1': 0.6110154905335629, 'number': 584}   | 0.6401            | 0.6336         | 0.6368     | 0.9125           |
| 0.0183        | 37.0  | 666  | 0.4656          | {'precision': 0.6632478632478632, 'recall': 0.6643835616438356, 'f1': 0.6638152266894781, 'number': 584}   | {'precision': 0.6, 'recall': 0.601027397260274, 'f1': 0.6005132591958939, 'number': 584}                   | 0.6316            | 0.6327         | 0.6322     | 0.9131           |
| 0.0209        | 38.0  | 684  | 0.4754          | {'precision': 0.6649214659685864, 'recall': 0.6523972602739726, 'f1': 0.658599827139153, 'number': 584}    | {'precision': 0.6073298429319371, 'recall': 0.5958904109589042, 'f1': 0.6015557476231633, 'number': 584}   | 0.6361            | 0.6241         | 0.6301     | 0.9131           |
| 0.0166        | 39.0  | 702  | 0.4703          | {'precision': 0.6695352839931153, 'recall': 0.666095890410959, 'f1': 0.6678111587982833, 'number': 584}    | {'precision': 0.612736660929432, 'recall': 0.6095890410958904, 'f1': 0.6111587982832618, 'number': 584}    | 0.6411            | 0.6378         | 0.6395     | 0.9151           |
| 0.0152        | 40.0  | 720  | 0.4739          | {'precision': 0.6626712328767124, 'recall': 0.6626712328767124, 'f1': 0.6626712328767124, 'number': 584}   | {'precision': 0.6215753424657534, 'recall': 0.6215753424657534, 'f1': 0.6215753424657534, 'number': 584}   | 0.6421            | 0.6421         | 0.6421     | 0.9139           |
| 0.0173        | 41.0  | 738  | 0.4839          | {'precision': 0.6610738255033557, 'recall': 0.6746575342465754, 'f1': 0.6677966101694915, 'number': 584}   | {'precision': 0.6191275167785235, 'recall': 0.6318493150684932, 'f1': 0.6254237288135593, 'number': 584}   | 0.6401            | 0.6533         | 0.6466     | 0.9139           |
| 0.0162        | 42.0  | 756  | 0.4854          | {'precision': 0.6610455311973018, 'recall': 0.6712328767123288, 'f1': 0.6661002548853017, 'number': 584}   | {'precision': 0.6138279932546374, 'recall': 0.6232876712328768, 'f1': 0.6185216652506373, 'number': 584}   | 0.6374            | 0.6473         | 0.6423     | 0.9156           |
| 0.0186        | 43.0  | 774  | 0.4747          | {'precision': 0.666095890410959, 'recall': 0.666095890410959, 'f1': 0.666095890410959, 'number': 584}      | {'precision': 0.6061643835616438, 'recall': 0.6061643835616438, 'f1': 0.6061643835616438, 'number': 584}   | 0.6361            | 0.6361         | 0.6361     | 0.9156           |
| 0.0149        | 44.0  | 792  | 0.4920          | {'precision': 0.6695501730103807, 'recall': 0.6626712328767124, 'f1': 0.666092943201377, 'number': 584}    | {'precision': 0.6141868512110726, 'recall': 0.6078767123287672, 'f1': 0.6110154905335629, 'number': 584}   | 0.6419            | 0.6353         | 0.6386     | 0.9139           |
| 0.0126        | 45.0  | 810  | 0.4911          | {'precision': 0.6621392190152802, 'recall': 0.6678082191780822, 'f1': 0.6649616368286446, 'number': 584}   | {'precision': 0.6146010186757216, 'recall': 0.6198630136986302, 'f1': 0.6172208013640239, 'number': 584}   | 0.6384            | 0.6438         | 0.6411     | 0.9117           |
| 0.0142        | 46.0  | 828  | 0.4932          | {'precision': 0.671280276816609, 'recall': 0.6643835616438356, 'f1': 0.6678141135972462, 'number': 584}    | {'precision': 0.6228373702422145, 'recall': 0.6164383561643836, 'f1': 0.6196213425129088, 'number': 584}   | 0.6471            | 0.6404         | 0.6437     | 0.9123           |
| 0.0107        | 47.0  | 846  | 0.5057          | {'precision': 0.6730103806228374, 'recall': 0.666095890410959, 'f1': 0.6695352839931152, 'number': 584}    | {'precision': 0.6245674740484429, 'recall': 0.6181506849315068, 'f1': 0.6213425129087781, 'number': 584}   | 0.6488            | 0.6421         | 0.6454     | 0.9139           |
| 0.0127        | 48.0  | 864  | 0.5076          | {'precision': 0.6800699300699301, 'recall': 0.666095890410959, 'f1': 0.6730103806228375, 'number': 584}    | {'precision': 0.6293706293706294, 'recall': 0.6164383561643836, 'f1': 0.6228373702422145, 'number': 584}   | 0.6547            | 0.6413         | 0.6479     | 0.9156           |
| 0.0116        | 49.0  | 882  | 0.5185          | {'precision': 0.6759098786828422, 'recall': 0.6678082191780822, 'f1': 0.6718346253229973, 'number': 584}   | {'precision': 0.6291161178509532, 'recall': 0.6215753424657534, 'f1': 0.6253229974160206, 'number': 584}   | 0.6525            | 0.6447         | 0.6486     | 0.9148           |
| 0.0099        | 50.0  | 900  | 0.5142          | {'precision': 0.6764705882352942, 'recall': 0.6695205479452054, 'f1': 0.6729776247848538, 'number': 584}   | {'precision': 0.629757785467128, 'recall': 0.6232876712328768, 'f1': 0.6265060240963856, 'number': 584}    | 0.6531            | 0.6464         | 0.6497     | 0.9156           |


### Framework versions

- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0