umutarpayy/biyoloji11v_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: 0.1271
• Train Accuracy: 0.9554
• Validation Loss: 0.0891
• Validation Accuracy: 0.9656
• 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': 5498, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'warmup_steps': 610, '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
2.3986	0.3682	1.1051	0.6540	0
0.9983	0.6745	0.7413	0.7472	1
0.7225	0.7612	0.5206	0.8221	2
0.5699	0.8105	0.3931	0.8896	3
0.4412	0.8561	0.3026	0.9129	4
0.3507	0.8843	0.2141	0.9227	5
0.2870	0.9004	0.1924	0.9337	6
0.2314	0.9212	0.1339	0.9534	7
0.1952	0.9338	0.1239	0.9534	8
0.1692	0.9401	0.0969	0.9632	9
0.1415	0.9515	0.0932	0.9669	10
0.1271	0.9554	0.0891	0.9656	11

Framework versions

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