NusaBERT-base-POSP / README.md
w11wo's picture
End of training
b4e8074 verified
metadata
license: mit
base_model: LazarusNLP/NusaBERT-base
tags:
  - generated_from_trainer
datasets:
  - indonlu
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: NusaBERT-base-POSP
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: indonlu
          type: indonlu
          config: posp
          split: validation
          args: posp
        metrics:
          - name: Precision
            type: precision
            value: 0.9577443609022557
          - name: Recall
            type: recall
            value: 0.9577443609022557
          - name: F1
            type: f1
            value: 0.9577443609022557
          - name: Accuracy
            type: accuracy
            value: 0.9577443609022557

NusaBERT-base-POSP

This model is a fine-tuned version of LazarusNLP/NusaBERT-base on the indonlu dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1472
  • Precision: 0.9577
  • Recall: 0.9577
  • F1: 0.9577
  • Accuracy: 0.9577

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: 64
  • 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
No log 1.0 420 0.2680 0.9203 0.9203 0.9203 0.9203
0.6283 2.0 840 0.2017 0.9379 0.9379 0.9379 0.9379
0.218 3.0 1260 0.1785 0.9449 0.9449 0.9449 0.9449
0.1612 4.0 1680 0.1692 0.9490 0.9490 0.9490 0.9490
0.1393 5.0 2100 0.1577 0.9511 0.9511 0.9511 0.9511
0.1119 6.0 2520 0.1503 0.9539 0.9539 0.9539 0.9539
0.1119 7.0 2940 0.1499 0.9549 0.9549 0.9549 0.9549
0.0943 8.0 3360 0.1542 0.9547 0.9547 0.9547 0.9547
0.0824 9.0 3780 0.1517 0.9558 0.9558 0.9558 0.9558
0.0785 10.0 4200 0.1519 0.9557 0.9557 0.9557 0.9557

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.2.0+cu118
  • Datasets 2.17.1
  • Tokenizers 0.15.1