umutarpayy/turkce6_bert

umutarpayy/berturkce_turkce_6_v6

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: 0.5811
• Train Accuracy: 0.9333
• Validation Loss: 0.4950
• Validation Accuracy: 0.9344
• 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': 743, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'warmup_steps': 82, '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
3.5568	0.0742	3.4276	0.0984	0
3.2726	0.1792	3.0073	0.2131	1
2.7026	0.3142	2.3118	0.4590	2
2.0574	0.5108	1.7425	0.5574	3
1.6287	0.6242	1.3675	0.7377	4
1.2799	0.7267	1.0404	0.7951	5
1.0124	0.8025	0.8324	0.8279	6
0.8252	0.8617	0.6604	0.8852	7
0.6918	0.8925	0.5631	0.9180	8
0.6090	0.9217	0.5101	0.9344	9
0.5811	0.9333	0.4950	0.9344	10

Framework versions

• Transformers 4.47.1
• TensorFlow 2.18.0
• Datasets 3.2.0
• Tokenizers 0.21.0