metadata
license: mit
base_model: SCUT-DLVCLab/lilt-roberta-en-base
tags:
- generated_from_trainer
datasets:
- funsd-layoutlmv3
model-index:
- name: dokki-lilt
results: []
dokki-lilt
This model is a fine-tuned version of SCUT-DLVCLab/lilt-roberta-en-base on the funsd-layoutlmv3 dataset. It achieves the following results on the evaluation set:
- Loss: 1.2732
- Answer: {'precision': 0.859504132231405, 'recall': 0.8910648714810282, 'f1': 0.8750000000000001, 'number': 817}
- Header: {'precision': 0.6355140186915887, 'recall': 0.5714285714285714, 'f1': 0.6017699115044248, 'number': 119}
- Question: {'precision': 0.886672710788758, 'recall': 0.9080779944289693, 'f1': 0.8972477064220182, 'number': 1077}
- Overall Precision: 0.8624
- Overall Recall: 0.8813
- Overall F1: 0.8717
- Overall Accuracy: 0.8133
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: 5e-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
- training_steps: 600
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Answer | Header | Question | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
|---|---|---|---|---|---|---|---|---|---|---|
| 0.6927 | 5.26 | 100 | 0.7696 | {'precision': 0.7593582887700535, 'recall': 0.8690330477356181, 'f1': 0.8105022831050229, 'number': 817} | {'precision': 0.5166666666666667, 'recall': 0.2605042016806723, 'f1': 0.3463687150837989, 'number': 119} | {'precision': 0.8429973238180196, 'recall': 0.8774373259052924, 'f1': 0.8598726114649682, 'number': 1077} | 0.7968 | 0.8376 | 0.8167 | 0.7782 |
| 0.1598 | 10.53 | 200 | 1.0235 | {'precision': 0.8254504504504504, 'recall': 0.8971848225214198, 'f1': 0.859824046920821, 'number': 817} | {'precision': 0.6097560975609756, 'recall': 0.42016806722689076, 'f1': 0.4975124378109453, 'number': 119} | {'precision': 0.8744313011828936, 'recall': 0.8922934076137419, 'f1': 0.8832720588235293, 'number': 1077} | 0.8429 | 0.8664 | 0.8545 | 0.7863 |
| 0.0536 | 15.79 | 300 | 1.1157 | {'precision': 0.8597785977859779, 'recall': 0.8555691554467564, 'f1': 0.8576687116564417, 'number': 817} | {'precision': 0.5789473684210527, 'recall': 0.46218487394957986, 'f1': 0.514018691588785, 'number': 119} | {'precision': 0.8663793103448276, 'recall': 0.9331476323119777, 'f1': 0.8985248100134109, 'number': 1077} | 0.8506 | 0.8738 | 0.8620 | 0.8083 |
| 0.0223 | 21.05 | 400 | 1.2167 | {'precision': 0.8367117117117117, 'recall': 0.9094247246022031, 'f1': 0.8715542521994134, 'number': 817} | {'precision': 0.5149253731343284, 'recall': 0.5798319327731093, 'f1': 0.5454545454545455, 'number': 119} | {'precision': 0.8898148148148148, 'recall': 0.8922934076137419, 'f1': 0.8910523875753361, 'number': 1077} | 0.8435 | 0.8808 | 0.8617 | 0.8083 |
| 0.0094 | 26.32 | 500 | 1.2313 | {'precision': 0.8564760793465578, 'recall': 0.8984088127294981, 'f1': 0.8769414575866189, 'number': 817} | {'precision': 0.6534653465346535, 'recall': 0.5546218487394958, 'f1': 0.6000000000000001, 'number': 119} | {'precision': 0.8853790613718412, 'recall': 0.9108635097493036, 'f1': 0.8979405034324943, 'number': 1077} | 0.8621 | 0.8847 | 0.8733 | 0.8161 |
| 0.0051 | 31.58 | 600 | 1.2732 | {'precision': 0.859504132231405, 'recall': 0.8910648714810282, 'f1': 0.8750000000000001, 'number': 817} | {'precision': 0.6355140186915887, 'recall': 0.5714285714285714, 'f1': 0.6017699115044248, 'number': 119} | {'precision': 0.886672710788758, 'recall': 0.9080779944289693, 'f1': 0.8972477064220182, 'number': 1077} | 0.8624 | 0.8813 | 0.8717 | 0.8133 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1