eraydikyologlu/bert_11_v3

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: 0.1471
• Train Accuracy: 0.9541
• Validation Loss: 0.1006
• Validation Accuracy: 0.9652
• 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': 32778, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'warmup_steps': 3642, '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
3.7539	0.2817	1.7655	0.5463	0
1.4174	0.5984	1.0370	0.6759	1
0.9943	0.6874	0.7886	0.7444	2
0.7834	0.7491	0.5958	0.8060	3
0.6283	0.7972	0.4637	0.8536	4
0.5069	0.8352	0.3383	0.8925	5
0.4070	0.8695	0.2612	0.9191	6
0.3288	0.8946	0.2056	0.9368	7
0.2655	0.9138	0.1536	0.9510	8
0.2181	0.9281	0.1381	0.9537	9
0.1779	0.9421	0.1078	0.9640	10
0.1471	0.9541	0.1006	0.9652	11

Framework versions

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