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: 1.2479
• Answer: {'precision': 0.8644859813084113, 'recall': 0.9057527539779682, 'f1': 0.8846383741781233, 'number': 817}
• Header: {'precision': 0.6262626262626263, 'recall': 0.5210084033613446, 'f1': 0.5688073394495413, 'number': 119}
• Question: {'precision': 0.8877005347593583, 'recall': 0.924791086350975, 'f1': 0.9058663028649386, 'number': 1077}
• Overall Precision: 0.8657
• Overall Recall: 0.8932
• Overall F1: 0.8792
• 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: 500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Answer	Header	Question	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.4245	10.53	200	0.9942	{'precision': 0.8187845303867404, 'recall': 0.9069767441860465, 'f1': 0.8606271777003485, 'number': 817}	{'precision': 0.5178571428571429, 'recall': 0.48739495798319327, 'f1': 0.5021645021645021, 'number': 119}	{'precision': 0.8821396192203083, 'recall': 0.903435468895079, 'f1': 0.8926605504587157, 'number': 1077}	0.8358	0.8803	0.8575	0.8150
0.0366	21.05	400	1.2479	{'precision': 0.8644859813084113, 'recall': 0.9057527539779682, 'f1': 0.8846383741781233, 'number': 817}	{'precision': 0.6262626262626263, 'recall': 0.5210084033613446, 'f1': 0.5688073394495413, 'number': 119}	{'precision': 0.8877005347593583, 'recall': 0.924791086350975, 'f1': 0.9058663028649386, 'number': 1077}	0.8657	0.8932	0.8792	0.8133

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2