umutarpayy/fizik11_bert

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: 0.0640
• Train Accuracy: 0.9874
• Validation Loss: 0.0379
• Validation Accuracy: 0.9937
• 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': 'transformers.optimization_tf', 'class_name': 'WarmUp', 'config': {'initial_learning_rate': 3e-05, 'decay_schedule_fn': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 3e-05, 'decay_steps': 6372, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'warmup_steps': 708, 'power': 1.0, 'name': None}, 'registered_name': 'WarmUp'}, '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
2.9687	0.3102	1.4885	0.6265	0
1.2355	0.6692	0.8219	0.7556	1
0.8233	0.7548	0.5530	0.8254	2
0.6070	0.8155	0.3960	0.8677	3
0.4561	0.8602	0.2857	0.9090	4
0.3428	0.8961	0.1730	0.9566	5
0.2502	0.9276	0.1162	0.9714	6
0.1830	0.9500	0.0958	0.9725	7
0.1413	0.9600	0.0643	0.9841	8
0.1030	0.9740	0.0483	0.9905	9
0.0768	0.9825	0.0413	0.9926	10
0.0640	0.9874	0.0379	0.9937	11

Framework versions

• Transformers 4.48.3
• TensorFlow 2.18.0
• Datasets 3.4.1
• Tokenizers 0.21.1