This is a BioBERT-based model is fine-tuned to perform Named Entity Recognition for drug names and adverse drug effects.

This model classifies input tokens into one of five classes:

B-DRUG: beginning of a drug entity I-DRUG: within a drug entity B-EFFECT: beginning of an AE entity I-EFFECT: within an AE entity O: outside either of the above entities

BioBERT-finetuned-ner

This model is a fine-tuned version of Dinithi/BioBERT on ade_corpus_v2. It achieves the following results on the evaluation set:

• Loss: 0.1602
• Precision: 0.8136
• Recall: 0.8961
• F1: 0.8528
• Accuracy: 0.9524

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• 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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.1673	1.0	113	0.2197	0.7545	0.8573	0.8027	0.9334
0.174	2.0	226	0.1691	0.7820	0.8870	0.8312	0.9472
0.1832	3.0	339	0.1596	0.8043	0.8915	0.8457	0.9506
0.0327	4.0	452	0.1591	0.8068	0.8980	0.8500	0.9526
0.036	5.0	565	0.1602	0.8136	0.8961	0.8528	0.9524

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu121
• Tokenizers 0.20.3

Dataset used

Dataset: https://huggingface.co/datasets/ade_corpus_v2