metadata

license: mit
tags:
  - generated_from_trainer
datasets:
  - lextreme
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: roberta-base-mapa_fine-ner
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: lextreme
          type: lextreme
          config: mapa_fine
          split: test
          args: mapa_fine
        metrics:
          - name: Precision
            type: precision
            value: 0.7395134779750164
          - name: Recall
            type: recall
            value: 0.8236672524897481
          - name: F1
            type: f1
            value: 0.7793251576248873
          - name: Accuracy
            type: accuracy
            value: 0.991740752278482

roberta-base-mapa_fine-ner

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

Loss: 0.0401
Precision: 0.7395
Recall: 0.8237
F1: 0.7793
Accuracy: 0.9917

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: 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	Precision	Recall	F1	Accuracy
0.0877	1.0	1739	0.0495	0.6861	0.7595	0.7209	0.9903
0.0661	2.0	3478	0.0432	0.7278	0.8092	0.7663	0.9914
0.0633	3.0	5217	0.0403	0.7469	0.8128	0.7785	0.9919
0.059	4.0	6956	0.0401	0.7412	0.8196	0.7784	0.9918
0.063	5.0	8695	0.0400	0.7425	0.8200	0.7793	0.9918
0.0593	6.0	10434	0.0405	0.7332	0.8244	0.7761	0.9916
0.0595	7.0	12173	0.0400	0.7389	0.8222	0.7783	0.9917
0.0593	8.0	13912	0.0401	0.7390	0.8229	0.7787	0.9917
0.0594	9.0	15651	0.0402	0.7374	0.8240	0.7783	0.9917
0.0597	10.0	17390	0.0401	0.7395	0.8237	0.7793	0.9917

Framework versions

Transformers 4.26.0
Pytorch 1.13.1+cu117
Datasets 2.9.0
Tokenizers 0.13.2