GutBrainIE_NER_baseline

This model is a fine-tuned version of microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3083
• Precision: 0.6802
• Recall: 0.6483
• F1: 0.6639
• Accuracy: 0.9188

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	92	0.4493	0.5809	0.4528	0.5089	0.8836
No log	2.0	184	0.3428	0.6849	0.5128	0.5865	0.9108
No log	3.0	276	0.2942	0.6480	0.6204	0.6339	0.9174
No log	4.0	368	0.2903	0.6780	0.6199	0.6476	0.9197
No log	5.0	460	0.2982	0.7156	0.6128	0.6602	0.9208
0.3823	6.0	552	0.2849	0.6767	0.6483	0.6622	0.9198
0.3823	7.0	644	0.2990	0.6705	0.6368	0.6532	0.9176
0.3823	8.0	736	0.3068	0.6756	0.6450	0.6600	0.9184
0.3823	9.0	828	0.3061	0.6812	0.6466	0.6635	0.9190
0.3823	10.0	920	0.3083	0.6802	0.6483	0.6639	0.9188

Framework versions

• Transformers 4.48.3
• Pytorch 2.6.0+cu124
• Datasets 3.3.0
• Tokenizers 0.21.0