metadata
license: mit
base_model: dbmdz/bert-base-turkish-uncased
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: ayla_ozetler3006_bertuncased_2024
results: []
ayla_ozetler3006_bertuncased_2024
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: 0.2438
- Accuracy: 0.9167
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 |
|---|---|---|---|---|
| 1.8163 | 1.0 | 10 | 1.7453 | 0.2 |
| 1.5027 | 2.0 | 20 | 1.1183 | 0.6741 |
| 0.8506 | 3.0 | 30 | 0.5805 | 0.8130 |
| 0.4512 | 4.0 | 40 | 0.4143 | 0.8296 |
| 0.2935 | 5.0 | 50 | 0.3245 | 0.8889 |
| 0.2011 | 6.0 | 60 | 0.2905 | 0.8889 |
| 0.1404 | 7.0 | 70 | 0.2802 | 0.9 |
| 0.0942 | 8.0 | 80 | 0.2509 | 0.9259 |
| 0.0653 | 9.0 | 90 | 0.2402 | 0.9204 |
| 0.0536 | 10.0 | 100 | 0.2438 | 0.9167 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.0+cu121
- Datasets 2.17.1
- Tokenizers 0.15.0