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: 0.6894
Answer: {'precision': 0.7383073496659243, 'recall': 0.8115055079559363, 'f1': 0.7731778425655977, 'number': 817}
Header: {'precision': 0.09090909090909091, 'recall': 0.01680672268907563, 'f1': 0.028368794326241134, 'number': 119}
Question: {'precision': 0.7214170692431562, 'recall': 0.8319405756731661, 'f1': 0.7727468736524363, 'number': 1077}
Overall Precision: 0.7220
Overall Recall: 0.7755
Overall F1: 0.7478
Overall Accuracy: 0.7574

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: 2e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 8
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 100
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
1.3948	5.2632	100	0.6894	{'precision': 0.7383073496659243, 'recall': 0.8115055079559363, 'f1': 0.7731778425655977, 'number': 817}	{'precision': 0.09090909090909091, 'recall': 0.01680672268907563, 'f1': 0.028368794326241134, 'number': 119}	{'precision': 0.7214170692431562, 'recall': 0.8319405756731661, 'f1': 0.7727468736524363, 'number': 1077}	0.7220	0.7755	0.7478	0.7574
0.3743	10.5263	200	0.7426	{'precision': 0.7905982905982906, 'recall': 0.9057527539779682, 'f1': 0.8442669709070165, 'number': 817}	{'precision': 0.4691358024691358, 'recall': 0.31932773109243695, 'f1': 0.37999999999999995, 'number': 119}	{'precision': 0.8246205733558178, 'recall': 0.9080779944289693, 'f1': 0.8643393725143613, 'number': 1077}	0.7971	0.8723	0.8330	0.7807
0.1593	15.7895	300	0.8605	{'precision': 0.8204570184983678, 'recall': 0.9228886168910648, 'f1': 0.868663594470046, 'number': 817}	{'precision': 0.43125, 'recall': 0.5798319327731093, 'f1': 0.49462365591397855, 'number': 119}	{'precision': 0.8476104598737602, 'recall': 0.872794800371402, 'f1': 0.8600182982616651, 'number': 1077}	0.8058	0.8758	0.8393	0.7764
0.0816	21.0526	400	0.9635	{'precision': 0.8474970896391153, 'recall': 0.8910648714810282, 'f1': 0.8687350835322196, 'number': 817}	{'precision': 0.5490196078431373, 'recall': 0.47058823529411764, 'f1': 0.5067873303167422, 'number': 119}	{'precision': 0.8643919510061242, 'recall': 0.9173630454967502, 'f1': 0.8900900900900901, 'number': 1077}	0.8422	0.8803	0.8608	0.7885
0.0611	26.3158	500	0.9700	{'precision': 0.8224719101123595, 'recall': 0.8959608323133414, 'f1': 0.8576449912126538, 'number': 817}	{'precision': 0.5566037735849056, 'recall': 0.4957983193277311, 'f1': 0.5244444444444444, 'number': 119}	{'precision': 0.8679750223015165, 'recall': 0.903435468895079, 'f1': 0.8853503184713377, 'number': 1077}	0.8333	0.8763	0.8542	0.7819

Framework versions

Transformers 4.47.0
Pytorch 2.5.1+cpu
Datasets 3.1.0
Tokenizers 0.21.0

dharmik3005/lilt-en-funsd

lilt-en-funsd

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

dharmik3005/
lilt-en-funsd