uisikdag/bert-base-turkish-cased-7allV03-tr

umit_7allV3_full

Dataset:

https://www.kaggle.com/datasets/savasy/ttc4900

0:siyaset 1:dünya 2:ekonomi 3:kültür 4:saglik 5:spor 6:teknoloji

This model is a fine-tuned version of dbmdz/bert-base-turkish-cased on given dataset. It achieves the following results on the evaluation set:

• Loss: 0.2934
• Accuracy: 0.9293

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.2867	1.0	214	0.2880	0.9136
0.1672	2.0	428	0.2658	0.9327
0.0735	3.0	642	0.2934	0.9293

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu117
• Datasets 2.10.1
• Tokenizers 0.13.2