indobert-model-ner
This model is a fine-tuned version of indolem/indobert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2296
- Precision: 0.8307
- Recall: 0.8454
- F1: 0.8380
- Accuracy: 0.9530
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: 16
- 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 |
|---|---|---|---|---|---|---|---|
| 0.4855 | 1.0 | 784 | 0.1729 | 0.8069 | 0.8389 | 0.8226 | 0.9499 |
| 0.1513 | 2.0 | 1568 | 0.1781 | 0.8086 | 0.8371 | 0.8226 | 0.9497 |
| 0.1106 | 3.0 | 2352 | 0.1798 | 0.8231 | 0.8475 | 0.8351 | 0.9531 |
| 0.0784 | 4.0 | 3136 | 0.1941 | 0.8270 | 0.8442 | 0.8355 | 0.9535 |
| 0.0636 | 5.0 | 3920 | 0.2085 | 0.8269 | 0.8514 | 0.8389 | 0.9548 |
| 0.0451 | 6.0 | 4704 | 0.2296 | 0.8307 | 0.8454 | 0.8380 | 0.9530 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2