eraydikyologlu/bert_ayt_kimya_hyperparameterTuned

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: 0.0817
• Train Accuracy: 0.9861
• Validation Loss: 0.0367
• Validation Accuracy: 0.9974
• Epoch: 24

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': 2e-05, 'decay_schedule_fn': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 4253, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'warmup_steps': 472, '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
4.1330	0.0432	3.7180	0.1632	0
3.0233	0.2965	2.1268	0.5061	1
1.7964	0.5563	1.2788	0.6545	2
1.2141	0.6855	0.9450	0.7387	3
0.9498	0.7448	0.7580	0.7951	4
0.7899	0.7855	0.6220	0.8212	5
0.6734	0.8163	0.4992	0.8672	6
0.5766	0.8430	0.4339	0.8863	7
0.4936	0.8689	0.3426	0.9167	8
0.4321	0.8844	0.2815	0.9288	9
0.3679	0.9032	0.2262	0.9444	10
0.3170	0.9192	0.1904	0.9549	11
0.2665	0.9330	0.1693	0.9592	12
0.2348	0.9435	0.1187	0.9792	13
0.2020	0.9506	0.1065	0.9818	14
0.1767	0.9591	0.0883	0.9852	15
0.1580	0.9636	0.0746	0.9878	16
0.1374	0.9707	0.0677	0.9887	17
0.1251	0.9739	0.0531	0.9922	18
0.1142	0.9765	0.0500	0.9948	19
0.1023	0.9802	0.0451	0.9965	20
0.0950	0.9823	0.0425	0.9957	21
0.0882	0.9838	0.0393	0.9974	22
0.0843	0.9850	0.0373	0.9974	23
0.0817	0.9861	0.0367	0.9974	24

Framework versions

• Transformers 4.52.4
• TensorFlow 2.18.0
• Datasets 2.14.4
• Tokenizers 0.21.1