anvorja/xlm-roberta-large-finetuned-sp-ner-mama-biomedical

xlm-roberta-large-finetuned-conll03-english-finetuned-ner-biomedical-spanish

This model is a fine-tuned version of FacebookAI/xlm-roberta-large-finetuned-conll03-english on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1526
• Precision: 0.8568
• Recall: 0.8258
• F1: 0.8410
• Accuracy: 0.9542

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• 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
• lr_scheduler_warmup_steps: 200
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	379	0.8877	0.5421	0.4232	0.4754	0.7697
0.8712	2.0	758	0.7159	0.5625	0.4761	0.5157	0.8265
0.1507	3.0	1137	0.4917	0.6528	0.5265	0.5829	0.8724
0.0984	4.0	1516	0.3969	0.7123	0.6516	0.6806	0.9005
0.0984	5.0	1895	0.3112	0.7463	0.6452	0.6920	0.9090
0.0732	6.0	2274	0.2653	0.8166	0.7239	0.7674	0.9299
0.0561	7.0	2653	0.2200	0.8006	0.7148	0.7553	0.9308
0.0465	8.0	3032	0.1590	0.8451	0.7884	0.8158	0.9485
0.0465	9.0	3411	0.1526	0.8568	0.8258	0.8410	0.9542
0.0396	10.0	3790	0.1494	0.8493	0.8142	0.8314	0.9526

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3