XLNet-base_LeNER-Br / README.md
CassioBN's picture
Update README.md
fe93910 verified
metadata
license: mit
base_model: xlnet/xlnet-base-cased
tags:
  - generated_from_trainer
datasets:
  - lener_br
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: XLNet-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: 0.8062054933875891
          - name: Recall
            type: recall
            value: 0.872317006053935
          - name: F1
            type: f1
            value: 0.8379592915675389
          - name: Accuracy
            type: accuracy
            value: 0.9783680282796544

XLNet-base_LeNER-Br

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

  • Loss: nan
  • Precision: 0.8062
  • Recall: 0.8723
  • F1: 0.8380
  • Accuracy: 0.9784

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
0.2531 1.0 979 nan 0.6037 0.7788 0.6801 0.9602
0.0531 2.0 1958 nan 0.6865 0.8184 0.7467 0.9657
0.0344 3.0 2937 nan 0.7079 0.8321 0.7650 0.9697
0.0214 4.0 3916 nan 0.7739 0.8514 0.8108 0.9765
0.0176 5.0 4895 nan 0.7407 0.8520 0.7924 0.9712
0.0109 6.0 5874 nan 0.7984 0.8696 0.8325 0.9773
0.0093 7.0 6853 nan 0.7944 0.8657 0.8285 0.9778
0.0056 8.0 7832 nan 0.8130 0.8756 0.8431 0.9779
0.0041 9.0 8811 nan 0.8171 0.8751 0.8451 0.9781
0.0034 10.0 9790 nan 0.8062 0.8723 0.8380 0.9784

Testing results

metrics={'test_loss': 0.10678809881210327, 'test_precision': 0.8132832080200502, 'test_recall': 0.8670674682698731, 'test_f1': 0.8393145813126414, 'test_accuracy': 0.9862667593953853, 'test_runtime': 42.9969, 'test_samples_per_second': 32.328, 'test_steps_per_second': 4.047})

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.3.0+cu121
  • Datasets 2.20.0
  • Tokenizers 0.19.1