czert_lr2e-05_bs4_train30

This model is a fine-tuned version of UWB-AIR/Czert-B-base-cased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3470
• Precision: 0.8193
• Recall: 0.8425
• F1: 0.8307
• Accuracy: 0.9068

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use 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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	8	1.2115	0.4125	0.1207	0.1868	0.6170
No log	2.0	16	0.9203	0.5928	0.3332	0.4266	0.7072
No log	3.0	24	0.7010	0.6317	0.5939	0.6122	0.8031
No log	4.0	32	0.5577	0.7048	0.6746	0.6894	0.8376
No log	5.0	40	0.4754	0.7480	0.7238	0.7357	0.8606
No log	6.0	48	0.4327	0.7784	0.7446	0.7611	0.8730
No log	7.0	56	0.3963	0.7933	0.7803	0.7868	0.8866
No log	8.0	64	0.3723	0.8041	0.8069	0.8055	0.8951
No log	9.0	72	0.3828	0.8135	0.7813	0.7970	0.8910
No log	10.0	80	0.3623	0.8097	0.8199	0.8148	0.9000
No log	11.0	88	0.3616	0.8339	0.8098	0.8217	0.9016
No log	12.0	96	0.3601	0.8202	0.8238	0.8220	0.9031
No log	13.0	104	0.3696	0.8170	0.8194	0.8182	0.9014
No log	14.0	112	0.3637	0.8394	0.8252	0.8322	0.9077
No log	15.0	120	0.3673	0.8329	0.8353	0.8341	0.9092

Framework versions

• Transformers 4.51.3
• Pytorch 2.7.0+cu126
• Datasets 3.6.0
• Tokenizers 0.21.1