eraydikyologlu/bert_11

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: 0.1517
• Train Accuracy: 0.9521
• Validation Loss: 0.1014
• Validation Accuracy: 0.9672
• 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': 32692, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'warmup_steps': 3632, '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
3.7943	0.2677	1.7506	0.5493	0
1.4352	0.5960	1.0051	0.6862	1
1.0035	0.6857	0.7639	0.7504	2
0.7961	0.7443	0.5834	0.8031	3
0.6388	0.7950	0.4540	0.8520	4
0.5128	0.8353	0.3429	0.8972	5
0.4100	0.8694	0.2714	0.9160	6
0.3308	0.8945	0.2011	0.9356	7
0.2703	0.9130	0.1606	0.9492	8
0.2208	0.9287	0.1278	0.9573	9
0.1796	0.9413	0.1106	0.9643	10
0.1517	0.9521	0.1014	0.9672	11

Framework versions

• Transformers 4.48.3
• TensorFlow 2.18.0
• Datasets 3.3.2
• Tokenizers 0.21.0