umutarpayy/matematik8_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: 0.0591
• Train Accuracy: 0.9881
• Validation Loss: 0.0292
• Validation Accuracy: 0.9932
• Epoch: 12

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': 12987, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'warmup_steps': 1443, '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
3.3665	0.1691	2.2028	0.4164	0
1.7677	0.5044	1.2978	0.6252	1
1.2807	0.6186	0.9497	0.7119	2
0.9818	0.6990	0.6892	0.7907	3
0.7449	0.7704	0.4490	0.8779	4
0.5489	0.8347	0.3160	0.9066	5
0.3926	0.8871	0.1755	0.9567	6
0.2796	0.9240	0.1192	0.9702	7
0.1945	0.9480	0.0745	0.9837	8
0.1414	0.9626	0.0593	0.9842	9
0.1047	0.9745	0.0438	0.9893	10
0.0778	0.9814	0.0327	0.9927	11
0.0591	0.9881	0.0292	0.9932	12

Framework versions

• Transformers 4.50.0
• TensorFlow 2.18.0
• Datasets 3.5.0
• Tokenizers 0.21.1