ner-biomedical-maccrobat2018

This model is a fine-tuned version of d4data/biomedical-ner-all on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6342
• Accuracy: 0.7903

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: 0.0001
• 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.5437	1.0	10	1.6009	0.3671
1.3262	2.0	20	0.9660	0.6254
0.8675	3.0	30	0.7436	0.7145
0.6199	4.0	40	0.6544	0.7385
0.4707	5.0	50	0.6131	0.7660
0.3735	6.0	60	0.6027	0.7709
0.3049	7.0	70	0.6056	0.7786
0.2507	8.0	80	0.5992	0.7792
0.2144	9.0	90	0.6115	0.7780
0.1801	10.0	100	0.6062	0.7863
0.1539	11.0	110	0.6101	0.7854
0.1372	12.0	120	0.6157	0.7892
0.1234	13.0	130	0.6269	0.7896
0.1119	14.0	140	0.6285	0.7881
0.1025	15.0	150	0.6364	0.7879
0.0945	16.0	160	0.6326	0.7896
0.09	17.0	170	0.6297	0.7916
0.0861	18.0	180	0.6318	0.7908
0.083	19.0	190	0.6317	0.7901
0.0817	20.0	200	0.6342	0.7903

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0