BerTurkBase_15_epoch

This model is a fine-tuned version of dbmdz/bert-base-turkish-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0005
• Accuracy: 1.0

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:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	50	0.6526	0.5972
No log	2.0	100	0.1755	0.9653
No log	3.0	150	0.0518	0.9861
No log	4.0	200	0.0065	1.0
No log	5.0	250	0.0022	1.0
No log	6.0	300	0.0016	1.0
No log	7.0	350	0.0007	1.0
No log	8.0	400	0.0005	1.0
No log	9.0	450	0.0005	1.0
0.1362	10.0	500	0.0005	1.0
0.1362	11.0	550	0.0006	1.0
0.1362	12.0	600	0.0005	1.0
0.1362	13.0	650	0.0005	1.0
0.1362	14.0	700	0.0005	1.0
0.1362	15.0	750	0.0005	1.0

Framework versions

• Transformers 4.25.1
• Pytorch 1.13.0+cu116
• Datasets 2.8.0
• Tokenizers 0.13.2