Bio_ClinicalBERT_fold_9_ternary_v1

This model is a fine-tuned version of emilyalsentzer/Bio_ClinicalBERT on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.0189
• F1: 0.7905

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	292	0.5758	0.7571
0.5482	2.0	584	0.6282	0.7609
0.5482	3.0	876	0.6823	0.7841
0.2346	4.0	1168	0.9898	0.7776
0.2346	5.0	1460	1.1397	0.7866
0.1001	6.0	1752	1.3832	0.7751
0.0447	7.0	2044	1.6002	0.7674
0.0447	8.0	2336	1.7265	0.7584
0.0171	9.0	2628	1.6650	0.7699
0.0171	10.0	2920	1.7322	0.7661
0.0156	11.0	3212	1.8071	0.7789
0.012	12.0	3504	1.8322	0.7841
0.012	13.0	3796	1.8948	0.7763
0.01	14.0	4088	1.7667	0.7918
0.01	15.0	4380	1.8538	0.7879
0.0063	16.0	4672	1.9763	0.7776
0.0063	17.0	4964	1.9970	0.7841
0.0028	18.0	5256	1.9366	0.7931
0.0003	19.0	5548	1.9709	0.7892
0.0003	20.0	5840	1.9460	0.7879
0.0044	21.0	6132	2.0280	0.7866
0.0044	22.0	6424	1.9423	0.7918
0.0013	23.0	6716	1.9618	0.7918
0.004	24.0	7008	2.0241	0.7905
0.004	25.0	7300	2.0189	0.7905

Framework versions

• Transformers 4.21.1
• Pytorch 1.12.0+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1