metadata
license: mit
tags:
- generated_from_trainer
datasets:
- funsd-layoutlmv3
model-index:
- name: lilt-en-funsd
results: []
lilt-en-funsd
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: 0.9151
- Answer: {'precision': 0.8149779735682819, 'recall': 0.9057527539779682, 'f1': 0.8579710144927536, 'number': 817}
- Header: {'precision': 0.49523809523809526, 'recall': 0.4369747899159664, 'f1': 0.4642857142857143, 'number': 119}
- Question: {'precision': 0.8627272727272727, 'recall': 0.8811513463324049, 'f1': 0.8718419843821773, 'number': 1077}
- Overall Precision: 0.8239
- Overall Recall: 0.8649
- Overall F1: 0.8439
- Overall Accuracy: 0.7891
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: 200
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Answer | Header | Question | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
|---|---|---|---|---|---|---|---|---|---|---|
| 0.7154 | 5.26 | 100 | 0.7542 | {'precision': 0.8251173708920188, 'recall': 0.8604651162790697, 'f1': 0.8424206111443978, 'number': 817} | {'precision': 0.45054945054945056, 'recall': 0.3445378151260504, 'f1': 0.3904761904761904, 'number': 119} | {'precision': 0.8157248157248157, 'recall': 0.924791086350975, 'f1': 0.866840731070496, 'number': 1077} | 0.8041 | 0.8644 | 0.8331 | 0.7915 |
| 0.1665 | 10.53 | 200 | 0.9151 | {'precision': 0.8149779735682819, 'recall': 0.9057527539779682, 'f1': 0.8579710144927536, 'number': 817} | {'precision': 0.49523809523809526, 'recall': 0.4369747899159664, 'f1': 0.4642857142857143, 'number': 119} | {'precision': 0.8627272727272727, 'recall': 0.8811513463324049, 'f1': 0.8718419843821773, 'number': 1077} | 0.8239 | 0.8649 | 0.8439 | 0.7891 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2