ardauzunoglu/berturk-base-for-tutorial-orderliness-classification

berturk-base-for-ordered-classification

This model is a fine-tuned version of dbmdz/bert-base-turkish-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4201
• Accuracy: {'accuracy': 0.8592592592592593}
• F1: {'f1': 0.87248322147651}
• Precision: {'precision': 0.8333333333333334}
• Recall: {'recall': 0.9154929577464789}

It achieves the following results on the test set:

• Accuracy: {'accuracy': 0.8667}
• F1: {'f1': 0.8767}
• Precision: {'precision': 0.8534}
• Recall: {'recall': 0.9014}

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.5255	1.0	40	0.4744	{'accuracy': 0.7851851851851852}	{'f1': 0.8176100628930818}	{'precision': 0.7386363636363636}	{'recall': 0.9154929577464789}
0.3723	2.0	80	0.3938	{'accuracy': 0.8666666666666667}	{'f1': 0.8767123287671234}	{'precision': 0.8533333333333334}	{'recall': 0.9014084507042254}
0.3527	3.0	120	0.4380	{'accuracy': 0.8296296296296296}	{'f1': 0.8496732026143791}	{'precision': 0.7926829268292683}	{'recall': 0.9154929577464789}
0.303	4.0	160	0.4216	{'accuracy': 0.8592592592592593}	{'f1': 0.87248322147651}	{'precision': 0.8333333333333334}	{'recall': 0.9154929577464789}
0.2825	5.0	200	0.4201	{'accuracy': 0.8592592592592593}	{'f1': 0.87248322147651}	{'precision': 0.8333333333333334}	{'recall': 0.9154929577464789}

Framework versions

• Transformers 4.25.1
• Pytorch 1.13.0+cu116
• Datasets 2.8.0
• Tokenizers 0.13.2