ayla_ozetler3006_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: 0.2677
- Accuracy: 0.9148
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 |
|---|---|---|---|---|
| 1.7743 | 1.0 | 10 | 1.6741 | 0.3389 |
| 1.3789 | 2.0 | 20 | 0.9867 | 0.6907 |
| 0.6919 | 3.0 | 30 | 0.4551 | 0.8278 |
| 0.364 | 4.0 | 40 | 0.3367 | 0.8778 |
| 0.2237 | 5.0 | 50 | 0.2699 | 0.8944 |
| 0.1481 | 6.0 | 60 | 0.3266 | 0.8667 |
| 0.1267 | 7.0 | 70 | 0.2515 | 0.9111 |
| 0.0726 | 8.0 | 80 | 0.2603 | 0.9167 |
| 0.0567 | 9.0 | 90 | 0.2595 | 0.9111 |
| 0.0461 | 10.0 | 100 | 0.2677 | 0.9148 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.11.0