biogpt-ner

This model is a fine-tuned version of microsoft/biogpt on the ncbi_disease dataset. It achieves the following results on the evaluation set:

• Loss: 0.1935
• Disease: {'precision': 0.5545851528384279, 'recall': 0.6195121951219512, 'f1': 0.5852534562211981, 'number': 1640}
• Overall Precision: 0.5546
• Overall Recall: 0.6195
• Overall F1: 0.5853
• Overall Accuracy: 0.9500

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Disease	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.3019	1.0	680	0.1758	{'precision': 0.46767617938264416, 'recall': 0.4896341463414634, 'f1': 0.4784033363121835, 'number': 1640}	0.4677	0.4896	0.4784	0.9394
0.1606	2.0	1360	0.1641	{'precision': 0.5137519460300985, 'recall': 0.6036585365853658, 'f1': 0.5550883095037846, 'number': 1640}	0.5138	0.6037	0.5551	0.9455
0.0743	3.0	2040	0.1935	{'precision': 0.5545851528384279, 'recall': 0.6195121951219512, 'f1': 0.5852534562211981, 'number': 1640}	0.5546	0.6195	0.5853	0.9500

Framework versions

• Transformers 4.51.2
• Pytorch 2.6.0
• Datasets 3.5.0
• Tokenizers 0.21.1