ayla_ozetler300_bertuncased
This model is a fine-tuned version of dbmdz/bert-base-turkish-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1056
- Accuracy: 0.9756
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: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 0.97 | 8 | 1.5103 | 0.48 |
| 1.5956 | 1.94 | 16 | 0.8089 | 0.7911 |
| 0.9875 | 2.91 | 24 | 0.3019 | 0.9289 |
| 0.3379 | 4.0 | 33 | 0.1606 | 0.9556 |
| 0.1349 | 4.97 | 41 | 0.1423 | 0.96 |
| 0.1349 | 5.94 | 49 | 0.1177 | 0.9667 |
| 0.0697 | 6.91 | 57 | 0.1122 | 0.9689 |
| 0.0434 | 8.0 | 66 | 0.1065 | 0.9756 |
| 0.0238 | 8.97 | 74 | 0.1060 | 0.9756 |
| 0.0288 | 9.7 | 80 | 0.1056 | 0.9756 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.11.0