metadata

library_name: transformers
license: mit
base_model: SCUT-DLVCLab/lilt-roberta-en-base
tags:
  - generated_from_trainer
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 an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.6340
Answer: {'precision': 0.8372352285395763, 'recall': 0.9192166462668299, 'f1': 0.8763127187864644, 'number': 817}
Header: {'precision': 0.5963302752293578, 'recall': 0.5462184873949579, 'f1': 0.5701754385964912, 'number': 119}
Question: {'precision': 0.9025974025974026, 'recall': 0.903435468895079, 'f1': 0.9030162412993039, 'number': 1077}
Overall Precision: 0.8584
Overall Recall: 0.8887
Overall F1: 0.8733
Overall Accuracy: 0.8024

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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
training_steps: 2500
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Answer	Header	Question	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.411	10.5263	200	0.9921	{'precision': 0.8178770949720671, 'recall': 0.8959608323133414, 'f1': 0.8551401869158878, 'number': 817}	{'precision': 0.5328467153284672, 'recall': 0.6134453781512605, 'f1': 0.5703125, 'number': 119}	{'precision': 0.8800738007380073, 'recall': 0.8857938718662952, 'f1': 0.882924571957427, 'number': 1077}	0.8313	0.8738	0.8520	0.7889
0.0439	21.0526	400	1.1727	{'precision': 0.8136511375947996, 'recall': 0.9192166462668299, 'f1': 0.8632183908045976, 'number': 817}	{'precision': 0.5728155339805825, 'recall': 0.4957983193277311, 'f1': 0.5315315315315315, 'number': 119}	{'precision': 0.8545135845749343, 'recall': 0.9052924791086351, 'f1': 0.87917042380523, 'number': 1077}	0.8237	0.8867	0.8541	0.7946
0.0137	31.5789	600	1.2732	{'precision': 0.8581730769230769, 'recall': 0.8739290085679314, 'f1': 0.865979381443299, 'number': 817}	{'precision': 0.5147058823529411, 'recall': 0.5882352941176471, 'f1': 0.5490196078431372, 'number': 119}	{'precision': 0.8467400508044031, 'recall': 0.9285051067780873, 'f1': 0.8857395925597875, 'number': 1077}	0.8302	0.8862	0.8573	0.7981
0.0075	42.1053	800	1.4152	{'precision': 0.8164627363737486, 'recall': 0.8984088127294981, 'f1': 0.8554778554778554, 'number': 817}	{'precision': 0.5897435897435898, 'recall': 0.5798319327731093, 'f1': 0.5847457627118645, 'number': 119}	{'precision': 0.8922934076137419, 'recall': 0.8922934076137419, 'f1': 0.8922934076137419, 'number': 1077}	0.8428	0.8763	0.8592	0.7992
0.0047	52.6316	1000	1.5824	{'precision': 0.8137254901960784, 'recall': 0.9143206854345165, 'f1': 0.8610951008645533, 'number': 817}	{'precision': 0.627906976744186, 'recall': 0.453781512605042, 'f1': 0.526829268292683, 'number': 119}	{'precision': 0.8995348837209303, 'recall': 0.8978644382544104, 'f1': 0.8986988847583643, 'number': 1077}	0.8504	0.8783	0.8641	0.7952
0.0018	63.1579	1200	1.6941	{'precision': 0.8627450980392157, 'recall': 0.9155446756425949, 'f1': 0.8883610451306414, 'number': 817}	{'precision': 0.5193798449612403, 'recall': 0.5630252100840336, 'f1': 0.5403225806451614, 'number': 119}	{'precision': 0.9080568720379147, 'recall': 0.8895078922934077, 'f1': 0.8986866791744841, 'number': 1077}	0.8645	0.8808	0.8725	0.7959
0.0017	73.6842	1400	1.6340	{'precision': 0.8372352285395763, 'recall': 0.9192166462668299, 'f1': 0.8763127187864644, 'number': 817}	{'precision': 0.5963302752293578, 'recall': 0.5462184873949579, 'f1': 0.5701754385964912, 'number': 119}	{'precision': 0.9025974025974026, 'recall': 0.903435468895079, 'f1': 0.9030162412993039, 'number': 1077}	0.8584	0.8887	0.8733	0.8024
0.0011	84.2105	1600	1.5738	{'precision': 0.8417508417508418, 'recall': 0.9179926560587516, 'f1': 0.8782201405152226, 'number': 817}	{'precision': 0.5545454545454546, 'recall': 0.5126050420168067, 'f1': 0.5327510917030567, 'number': 119}	{'precision': 0.8979779411764706, 'recall': 0.9071494893221913, 'f1': 0.902540415704388, 'number': 1077}	0.8559	0.8882	0.8718	0.8095
0.0005	94.7368	1800	1.5766	{'precision': 0.8384180790960452, 'recall': 0.9082007343941249, 'f1': 0.871915393654524, 'number': 817}	{'precision': 0.5714285714285714, 'recall': 0.47058823529411764, 'f1': 0.5161290322580646, 'number': 119}	{'precision': 0.8916211293260473, 'recall': 0.9090064995357474, 'f1': 0.9002298850574713, 'number': 1077}	0.8539	0.8828	0.8681	0.8108
0.0005	105.2632	2000	1.7020	{'precision': 0.8291347207009858, 'recall': 0.9265605875152999, 'f1': 0.8751445086705203, 'number': 817}	{'precision': 0.5648148148148148, 'recall': 0.5126050420168067, 'f1': 0.5374449339207047, 'number': 119}	{'precision': 0.8938134810710988, 'recall': 0.8987929433611885, 'f1': 0.8962962962962964, 'number': 1077}	0.8489	0.8872	0.8676	0.7844
0.0002	115.7895	2200	1.7386	{'precision': 0.8264192139737991, 'recall': 0.9265605875152999, 'f1': 0.8736295441431046, 'number': 817}	{'precision': 0.58, 'recall': 0.48739495798319327, 'f1': 0.5296803652968036, 'number': 119}	{'precision': 0.891643709825528, 'recall': 0.9015784586815228, 'f1': 0.8965835641735919, 'number': 1077}	0.8485	0.8872	0.8674	0.7826
0.0003	126.3158	2400	1.7405	{'precision': 0.8422818791946308, 'recall': 0.9216646266829865, 'f1': 0.880187025131502, 'number': 817}	{'precision': 0.5576923076923077, 'recall': 0.48739495798319327, 'f1': 0.5201793721973094, 'number': 119}	{'precision': 0.8932842686292548, 'recall': 0.9015784586815228, 'f1': 0.8974121996303143, 'number': 1077}	0.8547	0.8852	0.8697	0.7796

Framework versions

Transformers 4.47.1
Pytorch 2.5.1+cu121
Datasets 3.2.0
Tokenizers 0.21.0