vih_explainability3

This model is a fine-tuned version of PlanTL-GOB-ES/bsc-bio-ehr-es on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3951
• Roc Auc: 0.8213
• Ap Score: 0.7049
• Precision: 0.9836
• Recall: 0.6452
• F1: 0.7792

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Roc Auc	Ap Score	Precision	Recall	F1
0.4261	0.8475	100	0.3832	0.6129	0.3793	1.0	0.2258	0.3684
0.2405	1.6949	200	0.4736	0.6344	0.4138	1.0	0.2688	0.4237
0.2088	2.5424	300	0.3452	0.7729	0.6274	0.9808	0.5484	0.7034
0.2196	3.3898	400	0.3644	0.7151	0.5431	1.0	0.4301	0.6015
0.2068	4.2373	500	0.5156	0.6344	0.4138	1.0	0.2688	0.4237
0.1374	5.0847	600	0.3988	0.7944	0.6619	0.9821	0.5914	0.7383
0.1098	5.9322	700	0.3629	0.8051	0.6791	0.9828	0.6129	0.7550
0.0914	6.7797	800	0.3394	0.8240	0.6934	0.9531	0.6559	0.7771
0.088	7.6271	900	0.3612	0.8334	0.7009	0.9403	0.6774	0.7875
0.0787	8.4746	1000	0.3801	0.8213	0.7049	0.9836	0.6452	0.7792
0.0588	9.3220	1100	0.3951	0.8213	0.7049	0.9836	0.6452	0.7792

Framework versions

• Transformers 4.41.0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1