metadata
license: mit
base_model: dbmdz/bert-base-turkish-cased
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: results
results: []
results
This model is a fine-tuned version of dbmdz/bert-base-turkish-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1216
- Accuracy: 0.9767
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: 5e-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
- lr_scheduler_warmup_steps: 500
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.0625 | 0.12 | 200 | 0.3592 | 0.8799 |
| 0.2277 | 0.25 | 400 | 0.1771 | 0.9473 |
| 0.202 | 0.37 | 600 | 0.1594 | 0.9602 |
| 0.1711 | 0.5 | 800 | 0.1588 | 0.9612 |
| 0.1532 | 0.62 | 1000 | 0.1764 | 0.9525 |
| 0.1508 | 0.75 | 1200 | 0.1482 | 0.9627 |
| 0.1441 | 0.87 | 1400 | 0.1102 | 0.9719 |
| 0.1128 | 0.99 | 1600 | 0.1284 | 0.9686 |
| 0.07 | 1.12 | 1800 | 0.1321 | 0.9686 |
| 0.0812 | 1.24 | 2000 | 0.1365 | 0.9706 |
| 0.0858 | 1.37 | 2200 | 0.1046 | 0.9719 |
| 0.0636 | 1.49 | 2400 | 0.1263 | 0.9709 |
| 0.0861 | 1.62 | 2600 | 0.1081 | 0.9742 |
| 0.0623 | 1.74 | 2800 | 0.1148 | 0.9748 |
| 0.082 | 1.86 | 3000 | 0.0877 | 0.9778 |
| 0.0456 | 1.99 | 3200 | 0.1167 | 0.9751 |
| 0.0277 | 2.11 | 3400 | 0.1147 | 0.9759 |
| 0.0205 | 2.24 | 3600 | 0.1247 | 0.9753 |
| 0.0167 | 2.36 | 3800 | 0.1188 | 0.9776 |
| 0.0185 | 2.49 | 4000 | 0.1362 | 0.9762 |
| 0.0156 | 2.61 | 4200 | 0.1254 | 0.9762 |
| 0.0273 | 2.73 | 4400 | 0.1230 | 0.9768 |
| 0.019 | 2.86 | 4600 | 0.1245 | 0.9764 |
| 0.0102 | 2.98 | 4800 | 0.1216 | 0.9767 |
Framework versions
- Transformers 4.35.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.7
- Tokenizers 0.14.1