xlm-roberta-base-som-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: 0.3066
• F1: 0.3637
• Roc Auc: 0.6383
• Accuracy: 0.4788

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
0.3432	1.0	106	0.3461	0.0	0.5	0.3816
0.3258	2.0	212	0.3400	0.0	0.5	0.3816
0.3076	3.0	318	0.3131	0.0692	0.5230	0.4099
0.2959	4.0	424	0.3027	0.0745	0.5265	0.4046
0.2854	5.0	530	0.2920	0.2137	0.5712	0.4629
0.2715	6.0	636	0.2937	0.1736	0.5626	0.4576
0.233	7.0	742	0.3147	0.2084	0.5835	0.4594
0.2275	8.0	848	0.2829	0.2705	0.5963	0.4912
0.2052	9.0	954	0.2919	0.2695	0.6049	0.4735
0.186	10.0	1060	0.3022	0.2667	0.6142	0.4682
0.1805	11.0	1166	0.3008	0.3441	0.6314	0.4700
0.1808	12.0	1272	0.2973	0.3154	0.6202	0.4823
0.1516	13.0	1378	0.3045	0.3540	0.6412	0.4664
0.1532	14.0	1484	0.3053	0.3408	0.6302	0.4576
0.1466	15.0	1590	0.3000	0.3593	0.6400	0.4806
0.1373	16.0	1696	0.3056	0.3503	0.6358	0.4753
0.1343	17.0	1802	0.3054	0.3472	0.6336	0.4735
0.1326	18.0	1908	0.3053	0.3614	0.6376	0.4770
0.116	19.0	2014	0.3051	0.3635	0.6375	0.4788
0.1288	20.0	2120	0.3066	0.3637	0.6383	0.4788

Framework versions

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