lilt-en-funsd

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

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: 0.8047
• Answer: {'precision': 0.5882917466410749, 'recall': 0.7503059975520195, 'f1': 0.6594943518020442, 'number': 817}
• Header: {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}
• Question: {'precision': 0.629838142153413, 'recall': 0.8310120705663882, 'f1': 0.7165732586068856, 'number': 1077}
• Overall Precision: 0.6044
• Overall Recall: 0.7491
• Overall F1: 0.6690
• Overall Accuracy: 0.7169

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: 25

Training results

Framework versions

• Transformers 4.28.1
• Pytorch 1.13.0+cpu
• Datasets 2.11.0
• Tokenizers 0.13.3