README.md

metadata

library_name: transformers
base_model: FacebookAI/xlm-roberta-large-finetuned-conll03-english
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: xlm-roberta-large-clinical-ner-breast-cancer-sp
    results: []

xlm-roberta-large-clinical-ner-breast-cancer-sp

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: 0.2999
• Precision: 0.8965
• Recall: 0.8959
• F1: 0.8962
• Accuracy: 0.9474

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: 8
• total_train_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_ratio: 0.2
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
1.2687	1.0	213	1.3859	0.4556	0.3389	0.3887	0.6763
0.4857	2.0	426	0.5022	0.7673	0.7919	0.7794	0.8970
0.2519	3.0	639	0.3412	0.8407	0.8452	0.8430	0.9259
0.1671	4.0	852	0.3058	0.8711	0.8659	0.8685	0.9355
0.1423	5.0	1065	0.2983	0.8585	0.8659	0.8622	0.9340
0.0973	6.0	1278	0.2795	0.8773	0.8732	0.8753	0.9397
0.0655	7.0	1491	0.2775	0.8755	0.8726	0.8740	0.9393
0.0734	8.0	1704	0.2755	0.8799	0.8846	0.8822	0.9422
0.0575	9.0	1917	0.2900	0.8828	0.8793	0.8810	0.9409
0.0522	10.0	2130	0.2852	0.8864	0.8846	0.8855	0.9417
0.0559	11.0	2343	0.2735	0.8863	0.8893	0.8878	0.9441
0.0401	12.0	2556	0.2845	0.8833	0.8939	0.8886	0.9434
0.0326	13.0	2769	0.2845	0.8951	0.8933	0.8942	0.9462
0.0513	14.0	2982	0.2864	0.8886	0.8886	0.8886	0.9453
0.0223	15.0	3195	0.2920	0.8923	0.8899	0.8911	0.9455
0.0332	16.0	3408	0.2956	0.8906	0.8906	0.8906	0.9470
0.0262	17.0	3621	0.2987	0.8953	0.8959	0.8956	0.9469
0.018	18.0	3834	0.2999	0.8965	0.8959	0.8962	0.9474
0.02	19.0	4047	0.3023	0.8965	0.8959	0.8962	0.9472
0.0222	19.9088	4240	0.3023	0.8965	0.8959	0.8962	0.9474

Framework versions

• Transformers 4.48.2
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0