umutarpayy/bert_matematik_6

This model is a fine-tuned version of dbmdz/bert-base-turkish-cased on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.2283
• Train Accuracy: 0.9475
• Validation Loss: 0.1557
• Validation Accuracy: 0.9614
• Epoch: 11

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:

• optimizer: {'name': 'Adam', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 9300, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
• training_precision: float32

Training results

Train Loss	Train Accuracy	Validation Loss	Validation Accuracy	Epoch
3.0220	0.2760	1.7334	0.5717	0
1.5144	0.5985	1.1455	0.6699	1
1.1069	0.6872	0.8943	0.7295	2
0.8911	0.7434	0.6902	0.7931	3
0.7160	0.7937	0.5481	0.8486	4
0.5892	0.8327	0.4402	0.8744	5
0.4838	0.8665	0.3478	0.9010	6
0.4034	0.8944	0.2653	0.9300	7
0.3421	0.9069	0.2125	0.9436	8
0.2890	0.9263	0.1814	0.9549	9
0.2487	0.9384	0.1629	0.9589	10
0.2283	0.9475	0.1557	0.9614	11

Framework versions

• Transformers 4.47.1
• TensorFlow 2.17.1
• Datasets 3.2.0
• Tokenizers 0.21.0