metadata

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:

eval_loss: 1.7170
eval_ANSWER: {'precision': 0.8775510204081632, 'recall': 0.8947368421052632, 'f1': 0.886060606060606, 'number': 817}
eval_HEADER: {'precision': 0.4644808743169399, 'recall': 0.7142857142857143, 'f1': 0.5629139072847683, 'number': 119}
eval_QUESTION: {'precision': 0.8934348239771646, 'recall': 0.871866295264624, 'f1': 0.8825187969924814, 'number': 1077}
eval_overall_precision: 0.8491
eval_overall_recall: 0.8718
eval_overall_f1: 0.8603
eval_overall_accuracy: 0.7983
eval_runtime: 59.4689
eval_samples_per_second: 0.841
eval_steps_per_second: 0.118
epoch: 84.2105
step: 1600

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

Framework versions

Transformers 4.42.4
Pytorch 2.3.1+cpu
Datasets 2.20.0
Tokenizers 0.19.1