metadata

license: apache-2.0
base_model: PlanTL-GOB-ES/bsc-bio-ehr-es
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
model-index:
  - name: vih_explainability2
    results: []

vih_explainability2

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: 0.3811
Roc Auc: 0.8939
Ap Score: 0.7760
Precision: 0.9146
Recall: 0.8065
F1: 0.8571

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: 3e-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
0.3644	0.8475	100	0.2350	0.8402	0.6907	0.9028	0.6989	0.7879
0.1848	1.6949	200	0.2358	0.8765	0.7168	0.8588	0.7849	0.8202
0.1462	2.5424	300	0.2215	0.9021	0.7509	0.8571	0.8387	0.8478
0.1105	3.3898	400	0.2671	0.8778	0.7504	0.9114	0.7742	0.8372
0.079	4.2373	500	0.3124	0.8630	0.7338	0.92	0.7419	0.8214
0.0248	5.0847	600	0.3464	0.8765	0.7416	0.9	0.7742	0.8324
0.0127	5.9322	700	0.3822	0.8617	0.7248	0.9079	0.7419	0.8166
0.0089	6.7797	800	0.3625	0.8885	0.7674	0.9136	0.7957	0.8506
0.0042	7.6271	900	0.3643	0.8885	0.7674	0.9136	0.7957	0.8506
0.0052	8.4746	1000	0.3772	0.8939	0.7760	0.9146	0.8065	0.8571
0.0031	9.3220	1100	0.3811	0.8939	0.7760	0.9146	0.8065	0.8571

Framework versions

Transformers 4.41.0
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.19.1