umutarpayy/bert_mat-revize_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: 0.2012
• Train Accuracy: 0.9526
• Validation Loss: 0.1624
• Validation Accuracy: 0.9604
• 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': 9456, '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
2.6888	0.3561	1.5903	0.5769	0
1.4102	0.6143	1.0969	0.6838	1
1.0552	0.6969	0.8595	0.7623	2
0.8404	0.7595	0.6829	0.8138	3
0.6784	0.8082	0.5536	0.8479	4
0.5449	0.8484	0.4347	0.8827	5
0.4510	0.8782	0.3555	0.9025	6
0.3670	0.9019	0.2788	0.9255	7
0.3034	0.9201	0.2256	0.9382	8
0.2555	0.9369	0.1926	0.9485	9
0.2241	0.9458	0.1702	0.9580	10
0.2012	0.9526	0.1624	0.9604	11

Framework versions

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