roberta-base-biomedical-clinical-es

This model is a fine-tuned version of BSC-LT/roberta-base-biomedical-clinical-es on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3818
• Precision: 0.8662
• Recall: 0.8919
• F1: 0.8788
• Accuracy: 0.9366

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	280	0.3298	0.7938	0.8326	0.8127	0.9125
0.5032	2.0	560	0.2747	0.8433	0.8770	0.8598	0.9296
0.5032	3.0	840	0.2891	0.8499	0.8637	0.8567	0.9265
0.1441	4.0	1120	0.2790	0.8453	0.8904	0.8672	0.9338
0.1441	5.0	1400	0.3087	0.8519	0.8948	0.8728	0.9344
0.0841	6.0	1680	0.3025	0.8746	0.8993	0.8868	0.9383
0.0841	7.0	1960	0.3373	0.8656	0.8874	0.8764	0.9358
0.0481	8.0	2240	0.3364	0.8717	0.8859	0.8788	0.9372
0.0355	9.0	2520	0.3608	0.8576	0.8830	0.8701	0.9338
0.0355	10.0	2800	0.3800	0.8719	0.8874	0.8796	0.9344
0.0243	11.0	3080	0.3760	0.8637	0.8919	0.8776	0.9363
0.0243	12.0	3360	0.3718	0.8615	0.8844	0.8728	0.9349
0.0207	13.0	3640	0.3724	0.8644	0.8874	0.8757	0.9358
0.0207	14.0	3920	0.3848	0.8680	0.8963	0.8819	0.9372
0.017	15.0	4200	0.3818	0.8662	0.8919	0.8788	0.9366

Framework versions

• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1