Aleksandar's picture
add model
edeb914
|
raw
history blame
2.41 kB
metadata
tags:
  - generated_from_trainer
datasets:
  - null
metrics:
  - precision
  - recall
  - f1
  - accuracy
model_index:
  - name: bert-srb-ner-setimes
    results:
      - task:
          name: Token Classification
          type: token-classification
        metric:
          name: Accuracy
          type: accuracy
          value: 0.9632618605436434

bert-srb-ner-setimes

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1682
  • Precision: 0.8153
  • Recall: 0.8353
  • F1: 0.8251
  • Accuracy: 0.9633

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: 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
No log 1.0 207 0.2003 0.6991 0.7440 0.7209 0.9415
No log 2.0 414 0.1597 0.7202 0.7770 0.7475 0.9509
0.2342 3.0 621 0.1505 0.7619 0.8012 0.7811 0.9558
0.2342 4.0 828 0.1710 0.7822 0.8020 0.7920 0.9564
0.082 5.0 1035 0.1434 0.7969 0.8277 0.8120 0.9614
0.082 6.0 1242 0.1526 0.8001 0.8350 0.8171 0.9625
0.082 7.0 1449 0.1590 0.8061 0.8384 0.8219 0.9633
0.0353 8.0 1656 0.1616 0.8143 0.8296 0.8219 0.9634
0.0353 9.0 1863 0.1684 0.8167 0.8367 0.8265 0.9631
0.017 10.0 2070 0.1682 0.8153 0.8353 0.8251 0.9633

Framework versions

  • Transformers 4.9.2
  • Pytorch 1.9.0
  • Datasets 1.11.0
  • Tokenizers 0.10.1