Hugging Face's logo

nguyenkhoa2407
/
bert-base-cased-NER-favsbot-no-apostrophe-2022-11-07

Token Classification
Transformers
PyTorch
bert
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
bert-base-cased-NER-favsbot-no-apostrophe-2022-11-07 / README.md
nguyenkhoa2407's picture
nguyenkhoa2407
update model card README.md
67db813
preview code
|
raw
history blame contribute delete
3.79 kB
metadata
license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - favsbot
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: bert-base-cased-NER-favsbot-no-apostrophe-2022-11-07
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: favsbot
          type: favsbot
          config: default
          split: train
          args: default
        metrics:
          - name: Precision
            type: precision
            value: 0.8275862068965517
          - name: Recall
            type: recall
            value: 0.96
          - name: F1
            type: f1
            value: 0.888888888888889
          - name: Accuracy
            type: accuracy
            value: 0.9444444444444444

bert-base-cased-NER-favsbot-no-apostrophe-2022-11-07

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

  • Loss: 0.1169
  • Precision: 0.8276
  • Recall: 0.96
  • F1: 0.8889
  • Accuracy: 0.9444

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: 1.5e-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: 20

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 1.0 10 1.6302 0.0 0.0 0.0 0.5972
No log 2.0 20 1.0453 0.6667 0.08 0.1429 0.6389
No log 3.0 30 0.7286 0.8421 0.64 0.7273 0.8472
No log 4.0 40 0.5296 0.8 0.8 0.8000 0.8889
No log 5.0 50 0.3960 0.8214 0.92 0.8679 0.9306
No log 6.0 60 0.2987 0.8214 0.92 0.8679 0.9306
No log 7.0 70 0.2424 0.8276 0.96 0.8889 0.9444
No log 8.0 80 0.2151 0.8276 0.96 0.8889 0.9444
No log 9.0 90 0.1815 0.8276 0.96 0.8889 0.9444
No log 10.0 100 0.1675 0.8276 0.96 0.8889 0.9444
No log 11.0 110 0.1504 0.8276 0.96 0.8889 0.9444
No log 12.0 120 0.1410 0.8276 0.96 0.8889 0.9444
No log 13.0 130 0.1350 0.8276 0.96 0.8889 0.9444
No log 14.0 140 0.1281 0.8276 0.96 0.8889 0.9444
No log 15.0 150 0.1239 0.8276 0.96 0.8889 0.9444
No log 16.0 160 0.1190 0.8276 0.96 0.8889 0.9444
No log 17.0 170 0.1187 0.8276 0.96 0.8889 0.9444
No log 18.0 180 0.1180 0.8276 0.96 0.8889 0.9444
No log 19.0 190 0.1170 0.8276 0.96 0.8889 0.9444
No log 20.0 200 0.1169 0.8276 0.96 0.8889 0.9444

Framework versions

  • Transformers 4.21.1
  • Pytorch 1.12.1
  • Datasets 2.4.0
  • Tokenizers 0.12.1