metadata
license: mit
base_model: microsoft/layoutlm-base-uncased
tags:
- generated_from_trainer
model-index:
- name: layoutlm-custom_no_text
results: []
layoutlm-custom_no_text
This model is a fine-tuned version of microsoft/layoutlm-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1118
- Noise: {'precision': 0.8832116788321168, 'recall': 0.8832116788321168, 'f1': 0.8832116788321168, 'number': 548}
- Signal: {'precision': 0.8594890510948905, 'recall': 0.8594890510948905, 'f1': 0.8594890510948904, 'number': 548}
- Overall Precision: 0.8714
- Overall Recall: 0.8714
- Overall F1: 0.8714
- Overall Accuracy: 0.9773
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: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Noise | Signal | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
|---|---|---|---|---|---|---|---|---|---|
| 0.4739 | 1.0 | 18 | 0.1915 | {'precision': 0.6647398843930635, 'recall': 0.6295620437956204, 'f1': 0.6466729147141518, 'number': 548} | {'precision': 0.6782273603082851, 'recall': 0.6423357664233577, 'f1': 0.6597938144329897, 'number': 548} | 0.6715 | 0.6359 | 0.6532 | 0.9293 |
| 0.188 | 2.0 | 36 | 0.1127 | {'precision': 0.8265107212475633, 'recall': 0.7737226277372263, 'f1': 0.7992459943449576, 'number': 548} | {'precision': 0.7953216374269005, 'recall': 0.7445255474452555, 'f1': 0.769085768143261, 'number': 548} | 0.8109 | 0.7591 | 0.7842 | 0.9579 |
| 0.1052 | 3.0 | 54 | 0.0889 | {'precision': 0.8455743879472694, 'recall': 0.8193430656934306, 'f1': 0.8322520852641334, 'number': 548} | {'precision': 0.8248587570621468, 'recall': 0.7992700729927007, 'f1': 0.8118628359592215, 'number': 548} | 0.8352 | 0.8093 | 0.8221 | 0.9674 |
| 0.0645 | 4.0 | 72 | 0.0766 | {'precision': 0.8775510204081632, 'recall': 0.8631386861313869, 'f1': 0.8702851885924563, 'number': 548} | {'precision': 0.8552875695732839, 'recall': 0.8412408759124088, 'f1': 0.8482060717571298, 'number': 548} | 0.8664 | 0.8522 | 0.8592 | 0.9750 |
| 0.0427 | 5.0 | 90 | 0.0914 | {'precision': 0.8586956521739131, 'recall': 0.864963503649635, 'f1': 0.8618181818181818, 'number': 548} | {'precision': 0.8351449275362319, 'recall': 0.8412408759124088, 'f1': 0.8381818181818181, 'number': 548} | 0.8469 | 0.8531 | 0.8500 | 0.9730 |
| 0.0283 | 6.0 | 108 | 0.0987 | {'precision': 0.8756855575868373, 'recall': 0.8740875912408759, 'f1': 0.8748858447488584, 'number': 548} | {'precision': 0.8555758683729433, 'recall': 0.8540145985401459, 'f1': 0.8547945205479452, 'number': 548} | 0.8656 | 0.8641 | 0.8648 | 0.9761 |
| 0.0205 | 7.0 | 126 | 0.0988 | {'precision': 0.8646209386281588, 'recall': 0.8740875912408759, 'f1': 0.8693284936479129, 'number': 548} | {'precision': 0.8375451263537906, 'recall': 0.8467153284671532, 'f1': 0.8421052631578947, 'number': 548} | 0.8511 | 0.8604 | 0.8557 | 0.9742 |
| 0.0141 | 8.0 | 144 | 0.1086 | {'precision': 0.8706739526411658, 'recall': 0.8722627737226277, 'f1': 0.8714676390154968, 'number': 548} | {'precision': 0.8542805100182149, 'recall': 0.8558394160583942, 'f1': 0.8550592525068369, 'number': 548} | 0.8625 | 0.8641 | 0.8633 | 0.9753 |
| 0.012 | 9.0 | 162 | 0.1076 | {'precision': 0.8811700182815356, 'recall': 0.8795620437956204, 'f1': 0.8803652968036529, 'number': 548} | {'precision': 0.8592321755027422, 'recall': 0.8576642335766423, 'f1': 0.8584474885844748, 'number': 548} | 0.8702 | 0.8686 | 0.8694 | 0.9773 |
| 0.0104 | 10.0 | 180 | 0.1089 | {'precision': 0.8788990825688073, 'recall': 0.8740875912408759, 'f1': 0.8764867337602928, 'number': 548} | {'precision': 0.8568807339449541, 'recall': 0.8521897810218978, 'f1': 0.8545288197621226, 'number': 548} | 0.8679 | 0.8631 | 0.8655 | 0.9764 |
| 0.0101 | 11.0 | 198 | 0.1111 | {'precision': 0.8813868613138686, 'recall': 0.8813868613138686, 'f1': 0.8813868613138687, 'number': 548} | {'precision': 0.8594890510948905, 'recall': 0.8594890510948905, 'f1': 0.8594890510948904, 'number': 548} | 0.8704 | 0.8704 | 0.8704 | 0.9761 |
| 0.008 | 12.0 | 216 | 0.1049 | {'precision': 0.886654478976234, 'recall': 0.885036496350365, 'f1': 0.8858447488584474, 'number': 548} | {'precision': 0.8665447897623401, 'recall': 0.864963503649635, 'f1': 0.8657534246575344, 'number': 548} | 0.8766 | 0.875 | 0.8758 | 0.9778 |
| 0.0072 | 13.0 | 234 | 0.1094 | {'precision': 0.8775137111517367, 'recall': 0.8759124087591241, 'f1': 0.8767123287671232, 'number': 548} | {'precision': 0.8519195612431444, 'recall': 0.8503649635036497, 'f1': 0.8511415525114155, 'number': 548} | 0.8647 | 0.8631 | 0.8639 | 0.9759 |
| 0.007 | 14.0 | 252 | 0.1117 | {'precision': 0.8777372262773723, 'recall': 0.8777372262773723, 'f1': 0.8777372262773723, 'number': 548} | {'precision': 0.8540145985401459, 'recall': 0.8540145985401459, 'f1': 0.8540145985401459, 'number': 548} | 0.8659 | 0.8659 | 0.8659 | 0.9764 |
| 0.0084 | 15.0 | 270 | 0.1118 | {'precision': 0.8832116788321168, 'recall': 0.8832116788321168, 'f1': 0.8832116788321168, 'number': 548} | {'precision': 0.8594890510948905, 'recall': 0.8594890510948905, 'f1': 0.8594890510948904, 'number': 548} | 0.8714 | 0.8714 | 0.8714 | 0.9773 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0