kstn's picture
update model card README.md
c12e3d0
metadata
license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - id_nergrit_corpus
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: mobilebert-uncased-finetuned-ner
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: id_nergrit_corpus
          type: id_nergrit_corpus
          config: ner
          split: validation
          args: ner
        metrics:
          - name: Precision
            type: precision
            value: 0.6699979179679367
          - name: Recall
            type: recall
            value: 0.6136244458216141
          - name: F1
            type: f1
            value: 0.6405732911990843
          - name: Accuracy
            type: accuracy
            value: 0.8974442203210374

mobilebert-uncased-finetuned-ner

This model is a fine-tuned version of google/mobilebert-uncased on the id_nergrit_corpus dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3800
  • Precision: 0.6700
  • Recall: 0.6136
  • F1: 0.6406
  • Accuracy: 0.8974

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: 3

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.6239 1.0 1567 0.4989 0.5842 0.4877 0.5316 0.8688
0.5356 2.0 3134 0.4003 0.6368 0.5879 0.6113 0.8905
0.4035 3.0 4701 0.3800 0.6700 0.6136 0.6406 0.8974

Framework versions

  • Transformers 4.29.1
  • Pytorch 2.0.0+cu118
  • Datasets 2.12.0
  • Tokenizers 0.13.3