umutarpayy/fizik10_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: 0.0644
• Train Accuracy: 0.9819
• Validation Loss: 0.0297
• Validation Accuracy: 0.9920
• Epoch: 10

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': 5033, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'warmup_steps': 559, '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
2.7435	0.3030	1.1076	0.7015	0
0.8469	0.7674	0.5415	0.8474	1
0.5509	0.8385	0.3880	0.8876	2
0.4246	0.8746	0.2923	0.9157	3
0.3240	0.9033	0.2249	0.9290	4
0.2447	0.9257	0.1332	0.9639	5
0.1866	0.9418	0.0942	0.9719	6
0.1330	0.9608	0.0582	0.9880	7
0.1031	0.9702	0.0386	0.9933	8
0.0798	0.9765	0.0334	0.9946	9
0.0644	0.9819	0.0297	0.9920	10

Framework versions

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