README.md

metadata

license: mit
base_model: SCUT-DLVCLab/lilt-roberta-en-base
tags:
  - generated_from_trainer
model-index:
  - name: lilt-en-funsd-layoutlmv3
    results: []

lilt-en-funsd-layoutlmv3

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: 1.5778
• Answer: {'precision': 0.8764705882352941, 'recall': 0.9118727050183598, 'f1': 0.8938212357528493, 'number': 817}
• Header: {'precision': 0.6593406593406593, 'recall': 0.5042016806722689, 'f1': 0.5714285714285715, 'number': 119}
• Question: {'precision': 0.8901890189018902, 'recall': 0.9182915506035283, 'f1': 0.9040219378427787, 'number': 1077}
• Overall Precision: 0.8743
• Overall Recall: 0.8912
• Overall F1: 0.8827
• Overall Accuracy: 0.8242

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: 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
0.4133	10.5263	200	1.0790	{'precision': 0.8122941822173436, 'recall': 0.9057527539779682, 'f1': 0.8564814814814816, 'number': 817}	{'precision': 0.5463917525773195, 'recall': 0.44537815126050423, 'f1': 0.4907407407407407, 'number': 119}	{'precision': 0.8778280542986425, 'recall': 0.9006499535747446, 'f1': 0.8890925756186985, 'number': 1077}	0.8344	0.8758	0.8546	0.7789
0.0495	21.0526	400	1.1160	{'precision': 0.8738207547169812, 'recall': 0.9069767441860465, 'f1': 0.8900900900900901, 'number': 817}	{'precision': 0.6354166666666666, 'recall': 0.5126050420168067, 'f1': 0.5674418604651162, 'number': 119}	{'precision': 0.8720829732065687, 'recall': 0.9368616527390901, 'f1': 0.9033124440465533, 'number': 1077}	0.8620	0.8997	0.8804	0.8217
0.0138	31.5789	600	1.3117	{'precision': 0.8450867052023121, 'recall': 0.8947368421052632, 'f1': 0.8692033293697978, 'number': 817}	{'precision': 0.5348837209302325, 'recall': 0.5798319327731093, 'f1': 0.5564516129032259, 'number': 119}	{'precision': 0.8910433979686058, 'recall': 0.8960074280408542, 'f1': 0.8935185185185185, 'number': 1077}	0.8498	0.8768	0.8631	0.8064
0.0069	42.1053	800	1.5258	{'precision': 0.8543352601156069, 'recall': 0.9045287637698899, 'f1': 0.8787158145065399, 'number': 817}	{'precision': 0.5700934579439252, 'recall': 0.5126050420168067, 'f1': 0.5398230088495575, 'number': 119}	{'precision': 0.8849315068493151, 'recall': 0.8997214484679665, 'f1': 0.8922651933701657, 'number': 1077}	0.8558	0.8788	0.8672	0.8003
0.0035	52.6316	1000	1.5538	{'precision': 0.86, 'recall': 0.8947368421052632, 'f1': 0.8770245950809838, 'number': 817}	{'precision': 0.5737704918032787, 'recall': 0.5882352941176471, 'f1': 0.5809128630705394, 'number': 119}	{'precision': 0.8893023255813953, 'recall': 0.8876508820798514, 'f1': 0.8884758364312269, 'number': 1077}	0.8583	0.8728	0.8655	0.8103
0.0025	63.1579	1200	1.4460	{'precision': 0.8547297297297297, 'recall': 0.9290085679314566, 'f1': 0.8903225806451612, 'number': 817}	{'precision': 0.6483516483516484, 'recall': 0.4957983193277311, 'f1': 0.561904761904762, 'number': 119}	{'precision': 0.8898916967509025, 'recall': 0.9155060352831941, 'f1': 0.9025171624713959, 'number': 1077}	0.8644	0.8962	0.88	0.8193
0.0014	73.6842	1400	1.5030	{'precision': 0.8910648714810282, 'recall': 0.8910648714810282, 'f1': 0.8910648714810282, 'number': 817}	{'precision': 0.616822429906542, 'recall': 0.5546218487394958, 'f1': 0.5840707964601769, 'number': 119}	{'precision': 0.8925022583559169, 'recall': 0.9173630454967502, 'f1': 0.9047619047619048, 'number': 1077}	0.8774	0.8852	0.8813	0.8247
0.0011	84.2105	1600	1.5379	{'precision': 0.8779342723004695, 'recall': 0.9155446756425949, 'f1': 0.8963451168364289, 'number': 817}	{'precision': 0.6210526315789474, 'recall': 0.4957983193277311, 'f1': 0.5514018691588785, 'number': 119}	{'precision': 0.9005424954792043, 'recall': 0.924791086350975, 'f1': 0.9125057260650481, 'number': 1077}	0.8782	0.8957	0.8869	0.8281
0.0005	94.7368	1800	1.5872	{'precision': 0.8630609896432682, 'recall': 0.9179926560587516, 'f1': 0.8896797153024911, 'number': 817}	{'precision': 0.6333333333333333, 'recall': 0.4789915966386555, 'f1': 0.5454545454545454, 'number': 119}	{'precision': 0.8955495004541326, 'recall': 0.9155060352831941, 'f1': 0.9054178145087237, 'number': 1077}	0.8704	0.8907	0.8804	0.8211
0.0004	105.2632	2000	1.5457	{'precision': 0.8758782201405152, 'recall': 0.9155446756425949, 'f1': 0.8952722920406941, 'number': 817}	{'precision': 0.6703296703296703, 'recall': 0.5126050420168067, 'f1': 0.5809523809523809, 'number': 119}	{'precision': 0.8824577025823687, 'recall': 0.9201485608170845, 'f1': 0.9009090909090909, 'number': 1077}	0.8704	0.8942	0.8821	0.8231
0.0002	115.7895	2200	1.5761	{'precision': 0.8802816901408451, 'recall': 0.9179926560587516, 'f1': 0.8987417615338527, 'number': 817}	{'precision': 0.6630434782608695, 'recall': 0.5126050420168067, 'f1': 0.5781990521327014, 'number': 119}	{'precision': 0.8940754039497307, 'recall': 0.924791086350975, 'f1': 0.9091738931994524, 'number': 1077}	0.8780	0.8977	0.8877	0.8271
0.0002	126.3158	2400	1.5778	{'precision': 0.8764705882352941, 'recall': 0.9118727050183598, 'f1': 0.8938212357528493, 'number': 817}	{'precision': 0.6593406593406593, 'recall': 0.5042016806722689, 'f1': 0.5714285714285715, 'number': 119}	{'precision': 0.8901890189018902, 'recall': 0.9182915506035283, 'f1': 0.9040219378427787, 'number': 1077}	0.8743	0.8912	0.8827	0.8242

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.1+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1