xlmroberta-finetuned-sem_eval-rest14-english

This model is a fine-tuned version of xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0880
• F1: 0.4875
• Roc Auc: 0.8755
• Accuracy: 0.7087
• Hamming Loss: 0.0254

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy	Hamming Loss
No log	1.0	381	0.1858	0.0351	0.6366	0.2288	0.0609
0.2234	2.0	762	0.1354	0.1345	0.6905	0.3325	0.0418
0.1455	3.0	1143	0.1091	0.2639	0.7709	0.5138	0.0343
0.1064	4.0	1524	0.0971	0.3768	0.8313	0.6212	0.0282
0.1064	5.0	1905	0.0917	0.3767	0.8316	0.6312	0.0286
0.0807	6.0	2286	0.0880	0.3966	0.8459	0.65	0.0274
0.0657	7.0	2667	0.0922	0.3977	0.8513	0.6525	0.0283
0.0527	8.0	3048	0.0903	0.4129	0.8555	0.6575	0.028
0.0527	9.0	3429	0.0921	0.4298	0.8611	0.6713	0.0278
0.0453	10.0	3810	0.0882	0.4612	0.8697	0.6887	0.0261
0.0385	11.0	4191	0.0859	0.4567	0.8701	0.6925	0.0253
0.0339	12.0	4572	0.0881	0.4600	0.8726	0.7025	0.0258
0.0339	13.0	4953	0.0882	0.4818	0.8775	0.7063	0.0252
0.0304	14.0	5334	0.0879	0.4793	0.8742	0.7	0.0254
0.0271	15.0	5715	0.0880	0.4875	0.8755	0.7087	0.0254

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1