metadata

license: mit
base_model: FacebookAI/xlm-roberta-base
tags:
  - generated_from_trainer
datasets:
  - lener_br
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: xlm-roberta-base_LeNER-Br
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: lener_br
          type: lener_br
          config: lener_br
          split: validation
          args: lener_br
        metrics:
          - name: Precision
            type: precision
            value: 0.8295165394402035
          - name: Recall
            type: recall
            value: 0.8965896589658966
          - name: F1
            type: f1
            value: 0.8617499339148824
          - name: Accuracy
            type: accuracy
            value: 0.9714166181062949

xlm-roberta-base_LeNER-Br

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the lener_br dataset. It achieves the following results on the evaluation set:

Loss: nan
Precision: 0.8295
Recall: 0.8966
F1: 0.8617
Accuracy: 0.9714

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.2394	1.0	979	nan	0.7134	0.8614	0.7805	0.9638
0.0375	2.0	1958	nan	0.8035	0.9043	0.8509	0.9670
0.0256	3.0	2937	nan	0.8026	0.8878	0.8430	0.9761
0.0194	4.0	3916	nan	0.7836	0.8861	0.8317	0.9670
0.015	5.0	4895	nan	0.8061	0.8988	0.8499	0.9691
0.0098	6.0	5874	nan	0.8279	0.9076	0.8659	0.9715
0.0082	7.0	6853	nan	0.8067	0.8905	0.8465	0.9681
0.0042	8.0	7832	nan	0.8233	0.9021	0.8609	0.9737
0.0037	9.0	8811	nan	0.8281	0.9010	0.8630	0.9712
0.0031	10.0	9790	nan	0.8295	0.8966	0.8617	0.9714

Testing Results

metrics={'test_loss': 0.07461995631456375, 'test_precision': 0.8852040816326531, 'test_recall': 0.9137590520079, 'test_f1': 0.8992549400712667, 'test_accuracy': 0.9883402014967543, 'test_runtime': 13.0766, 'test_samples_per_second': 106.297, 'test_steps_per_second': 13.306})

Framework versions

Transformers 4.41.2
Pytorch 2.3.0+cu121
Datasets 2.20.0
Tokenizers 0.19.1