File size: 9,393 Bytes
ec78391 |
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 |
---
library_name: transformers
license: mit
base_model: microsoft/layoutlm-base-uncased
tags:
- generated_from_trainer
datasets:
- funsd
model-index:
- name: layoutlm-FUNSD-only-1fold
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-FUNSD-only-1fold
This model is a fine-tuned version of [microsoft/layoutlm-base-uncased](https://huggingface.co/microsoft/layoutlm-base-uncased) on the funsd dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6561
- Eader: {'precision': 0.3770491803278688, 'recall': 0.27710843373493976, 'f1': 0.3194444444444444, 'number': 83}
- Nswer: {'precision': 0.5213675213675214, 'recall': 0.5951219512195122, 'f1': 0.5558086560364465, 'number': 205}
- Uestion: {'precision': 0.3722627737226277, 'recall': 0.44155844155844154, 'f1': 0.40396039603960393, 'number': 231}
- Overall Precision: 0.4341
- Overall Recall: 0.4759
- Overall F1: 0.4540
- Overall Accuracy: 0.7927
## 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: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 15
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Eader | Nswer | Uestion | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:----------------------------------------------------------------------------------------------------------:|:----------------------------------------------------------------------------------------------------------:|:-----------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
| 1.2996 | 1.0 | 8 | 1.0787 | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 83} | {'precision': 0.0748502994011976, 'recall': 0.24390243902439024, 'f1': 0.11454753722794961, 'number': 205} | {'precision': 0.0704647676161919, 'recall': 0.20346320346320346, 'f1': 0.10467706013363029, 'number': 231} | 0.0727 | 0.1869 | 0.1046 | 0.6075 |
| 1.0224 | 2.0 | 16 | 0.8740 | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 83} | {'precision': 0.20823798627002288, 'recall': 0.44390243902439025, 'f1': 0.2834890965732087, 'number': 205} | {'precision': 0.17551963048498845, 'recall': 0.329004329004329, 'f1': 0.2289156626506024, 'number': 231} | 0.1920 | 0.3218 | 0.2405 | 0.7001 |
| 0.8193 | 3.0 | 24 | 0.6924 | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 83} | {'precision': 0.310126582278481, 'recall': 0.47804878048780486, 'f1': 0.3761996161228407, 'number': 205} | {'precision': 0.25671641791044775, 'recall': 0.3722943722943723, 'f1': 0.30388692579505294, 'number': 231} | 0.2759 | 0.3545 | 0.3103 | 0.7519 |
| 0.6764 | 4.0 | 32 | 0.6312 | {'precision': 0.14285714285714285, 'recall': 0.04819277108433735, 'f1': 0.07207207207207209, 'number': 83} | {'precision': 0.3914590747330961, 'recall': 0.5365853658536586, 'f1': 0.45267489711934156, 'number': 205} | {'precision': 0.36, 'recall': 0.42857142857142855, 'f1': 0.391304347826087, 'number': 231} | 0.3647 | 0.4104 | 0.3862 | 0.7634 |
| 0.5456 | 5.0 | 40 | 0.5856 | {'precision': 0.3, 'recall': 0.18072289156626506, 'f1': 0.2255639097744361, 'number': 83} | {'precision': 0.42424242424242425, 'recall': 0.5463414634146342, 'f1': 0.47761194029850745, 'number': 205} | {'precision': 0.38022813688212925, 'recall': 0.4329004329004329, 'f1': 0.4048582995951417, 'number': 231} | 0.3934 | 0.4374 | 0.4142 | 0.7846 |
| 0.4547 | 6.0 | 48 | 0.5899 | {'precision': 0.4074074074074074, 'recall': 0.26506024096385544, 'f1': 0.32116788321167883, 'number': 83} | {'precision': 0.44129554655870445, 'recall': 0.5317073170731708, 'f1': 0.4823008849557522, 'number': 205} | {'precision': 0.35094339622641507, 'recall': 0.4025974025974026, 'f1': 0.37500000000000006, 'number': 231} | 0.3958 | 0.4316 | 0.4129 | 0.7827 |
| 0.3815 | 7.0 | 56 | 0.5921 | {'precision': 0.3888888888888889, 'recall': 0.25301204819277107, 'f1': 0.3065693430656934, 'number': 83} | {'precision': 0.483739837398374, 'recall': 0.5804878048780487, 'f1': 0.5277161862527716, 'number': 205} | {'precision': 0.34657039711191334, 'recall': 0.4155844155844156, 'f1': 0.3779527559055118, 'number': 231} | 0.4090 | 0.4547 | 0.4307 | 0.7893 |
| 0.324 | 8.0 | 64 | 0.5872 | {'precision': 0.5, 'recall': 0.26506024096385544, 'f1': 0.3464566929133859, 'number': 83} | {'precision': 0.4957983193277311, 'recall': 0.5756097560975609, 'f1': 0.5327313769751693, 'number': 205} | {'precision': 0.39543726235741444, 'recall': 0.45021645021645024, 'f1': 0.42105263157894735, 'number': 231} | 0.4477 | 0.4701 | 0.4586 | 0.7989 |
| 0.284 | 9.0 | 72 | 0.6026 | {'precision': 0.3968253968253968, 'recall': 0.30120481927710846, 'f1': 0.34246575342465757, 'number': 83} | {'precision': 0.4717741935483871, 'recall': 0.5707317073170731, 'f1': 0.5165562913907285, 'number': 205} | {'precision': 0.35789473684210527, 'recall': 0.44155844155844154, 'f1': 0.39534883720930236, 'number': 231} | 0.4094 | 0.4701 | 0.4377 | 0.7897 |
| 0.249 | 10.0 | 80 | 0.6137 | {'precision': 0.4423076923076923, 'recall': 0.27710843373493976, 'f1': 0.34074074074074073, 'number': 83} | {'precision': 0.5041322314049587, 'recall': 0.5951219512195122, 'f1': 0.5458612975391499, 'number': 205} | {'precision': 0.3607142857142857, 'recall': 0.43722943722943725, 'f1': 0.39530332681017616, 'number': 231} | 0.4286 | 0.4740 | 0.4501 | 0.7981 |
| 0.2288 | 11.0 | 88 | 0.6367 | {'precision': 0.38571428571428573, 'recall': 0.3253012048192771, 'f1': 0.35294117647058826, 'number': 83} | {'precision': 0.48717948717948717, 'recall': 0.5560975609756098, 'f1': 0.5193621867881548, 'number': 205} | {'precision': 0.38267148014440433, 'recall': 0.4588744588744589, 'f1': 0.4173228346456693, 'number': 231} | 0.4251 | 0.4759 | 0.4491 | 0.7912 |
| 0.2031 | 12.0 | 96 | 0.6401 | {'precision': 0.46, 'recall': 0.27710843373493976, 'f1': 0.3458646616541353, 'number': 83} | {'precision': 0.497907949790795, 'recall': 0.5804878048780487, 'f1': 0.536036036036036, 'number': 205} | {'precision': 0.3800738007380074, 'recall': 0.4458874458874459, 'f1': 0.4103585657370518, 'number': 231} | 0.4375 | 0.4721 | 0.4541 | 0.7985 |
| 0.193 | 13.0 | 104 | 0.6539 | {'precision': 0.37142857142857144, 'recall': 0.3132530120481928, 'f1': 0.33986928104575165, 'number': 83} | {'precision': 0.5321100917431193, 'recall': 0.5658536585365853, 'f1': 0.5484633569739952, 'number': 205} | {'precision': 0.3969465648854962, 'recall': 0.45021645021645024, 'f1': 0.4219066937119675, 'number': 231} | 0.4473 | 0.4740 | 0.4602 | 0.7904 |
| 0.1895 | 14.0 | 112 | 0.6557 | {'precision': 0.39344262295081966, 'recall': 0.2891566265060241, 'f1': 0.3333333333333333, 'number': 83} | {'precision': 0.5429864253393665, 'recall': 0.5853658536585366, 'f1': 0.5633802816901408, 'number': 205} | {'precision': 0.3916349809885932, 'recall': 0.4458874458874459, 'f1': 0.41700404858299595, 'number': 231} | 0.4532 | 0.4759 | 0.4643 | 0.7904 |
| 0.1775 | 15.0 | 120 | 0.6561 | {'precision': 0.3770491803278688, 'recall': 0.27710843373493976, 'f1': 0.3194444444444444, 'number': 83} | {'precision': 0.5213675213675214, 'recall': 0.5951219512195122, 'f1': 0.5558086560364465, 'number': 205} | {'precision': 0.3722627737226277, 'recall': 0.44155844155844154, 'f1': 0.40396039603960393, 'number': 231} | 0.4341 | 0.4759 | 0.4540 | 0.7927 |
### Framework versions
- Transformers 4.49.0
- Pytorch 2.6.0+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0
|