turkish-hs-group-predictionturkish-hs-group-prediction

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

• Loss: 0.3707
• Accuracy: 0.9194
• Macro F1: 0.8311

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-06
• train_batch_size: 16
• eval_batch_size: 20
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro F1
2.0719	0.1391	100	1.7014	0.5055	0.2155
1.4734	0.2782	200	1.1549	0.7277	0.5053
1.0189	0.4172	300	0.7536	0.8740	0.7174
0.7469	0.5563	400	0.5380	0.8975	0.7470
0.5813	0.6954	500	0.4410	0.9085	0.7775
0.4769	0.8345	600	0.3854	0.9147	0.8039
0.4088	0.9736	700	0.3707	0.9194	0.8311

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1