metadata
license: mit
base_model: dbmdz/distilbert-base-turkish-cased
tags:
- generated_from_trainer
metrics:
- f1
- accuracy
model-index:
- name: acer_nitro_distilbert_turk
results: []
acer_nitro_distilbert_turk
This model is a fine-tuned version of dbmdz/distilbert-base-turkish-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7724
- F1: 0.7596
- Roc Auc: 0.8506
- Accuracy: 0.6386
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: 2
- eval_batch_size: 2
- 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 | F1 | Roc Auc | Accuracy |
|---|---|---|---|---|---|---|
| No log | 1.0 | 166 | 0.6408 | 0.7330 | 0.8464 | 0.5904 |
| No log | 2.0 | 332 | 0.6674 | 0.7549 | 0.8434 | 0.6386 |
| No log | 3.0 | 498 | 0.7058 | 0.7573 | 0.8470 | 0.6145 |
| 0.3537 | 4.0 | 664 | 0.7887 | 0.7333 | 0.8358 | 0.6145 |
| 0.3537 | 5.0 | 830 | 0.8267 | 0.7586 | 0.8447 | 0.6265 |
| 0.3537 | 6.0 | 996 | 0.7217 | 0.7477 | 0.8491 | 0.6506 |
| 0.191 | 7.0 | 1162 | 0.7180 | 0.7593 | 0.8588 | 0.6506 |
| 0.191 | 8.0 | 1328 | 0.7742 | 0.7670 | 0.8531 | 0.6386 |
| 0.191 | 9.0 | 1494 | 0.7798 | 0.7560 | 0.8493 | 0.6265 |
| 0.1294 | 10.0 | 1660 | 0.7724 | 0.7596 | 0.8506 | 0.6386 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0