xlm-roberta-base-yor-noaug

This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2598
• F1: 0.1897
• Roc Auc: 0.5687
• Accuracy: 0.5815

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: 32
• eval_batch_size: 32
• 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: cosine
• lr_scheduler_warmup_steps: 100
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.306	1.0	94	0.2789	0.0	0.5	0.4748
0.2912	2.0	188	0.2793	0.0	0.5	0.4748
0.2806	3.0	282	0.2786	0.0	0.5	0.4748
0.2821	4.0	376	0.2783	0.0	0.5	0.4748
0.2631	5.0	470	0.2674	0.0670	0.5193	0.5151
0.2622	6.0	564	0.2577	0.0963	0.5344	0.5392
0.2509	7.0	658	0.2567	0.0984	0.5361	0.5453
0.2416	8.0	752	0.2507	0.1441	0.5512	0.5775
0.2218	9.0	846	0.2514	0.1693	0.5608	0.5755
0.2216	10.0	940	0.2458	0.1722	0.5606	0.5734
0.1995	11.0	1034	0.2493	0.1787	0.5648	0.5875
0.2041	12.0	1128	0.2564	0.1872	0.5681	0.5875
0.1811	13.0	1222	0.2598	0.1897	0.5687	0.5815
0.1774	14.0	1316	0.2608	0.1777	0.5666	0.5875
0.1637	15.0	1410	0.2606	0.1894	0.5653	0.5714
0.1665	16.0	1504	0.2669	0.1890	0.5696	0.5855
0.1618	17.0	1598	0.2655	0.1834	0.5672	0.5775

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0