xlm-roberta-base-esp-finetuned

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: 0.3345
• F1: 0.7715
• Roc Auc: 0.8559
• Accuracy: 0.6033

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
0.556	1.0	98	0.4924	0.1173	0.5484	0.125
0.3942	2.0	196	0.3753	0.6280	0.7772	0.4293
0.3051	3.0	294	0.3283	0.7282	0.8250	0.5380
0.2566	4.0	392	0.3234	0.7277	0.8307	0.5380
0.2077	5.0	490	0.3109	0.7502	0.8392	0.5652
0.1646	6.0	588	0.3135	0.7383	0.8336	0.5435
0.1524	7.0	686	0.3132	0.7456	0.8359	0.5707
0.1346	8.0	784	0.3253	0.7427	0.8341	0.5380
0.1076	9.0	882	0.3272	0.7549	0.8457	0.5924
0.0963	10.0	980	0.3384	0.7671	0.8528	0.5978
0.0888	11.0	1078	0.3381	0.7620	0.8485	0.5870
0.0762	12.0	1176	0.3345	0.7715	0.8559	0.6033
0.0528	13.0	1274	0.3566	0.7683	0.8577	0.5924
0.0512	14.0	1372	0.3522	0.7643	0.8534	0.5924
0.0435	15.0	1470	0.3595	0.7635	0.8517	0.5978
0.0415	16.0	1568	0.3651	0.7646	0.8550	0.5870

Framework versions

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