results

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: 2.1290
• Accuracy: 0.7103
• Precision: 0.7115
• Recall: 0.7103
• F1: 0.7101

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	Accuracy	Precision	Recall	F1
1.0332	1.0	702	0.9235	0.6014	0.6068	0.6014	0.6028
0.9128	2.0	1404	0.8918	0.6512	0.6649	0.6512	0.6459
0.7072	3.0	2106	0.7823	0.6833	0.6934	0.6833	0.6847
0.6127	4.0	2808	0.8272	0.6989	0.7004	0.6989	0.6992
0.4895	5.0	3510	0.9571	0.6975	0.6999	0.6975	0.6963
0.4169	6.0	4212	1.0834	0.7196	0.7221	0.7196	0.7184
0.3969	7.0	4914	1.2723	0.7046	0.7097	0.7046	0.7041
0.3436	8.0	5616	1.6584	0.7032	0.7113	0.7032	0.7038
0.3081	9.0	6318	1.6509	0.6954	0.7049	0.6954	0.6942
0.2431	10.0	7020	1.8800	0.7082	0.7144	0.7082	0.7075
0.2428	11.0	7722	1.9124	0.7146	0.7165	0.7146	0.7133
0.178	12.0	8424	1.9060	0.7196	0.7209	0.7196	0.7191
0.1337	13.0	9126	2.0505	0.7032	0.7119	0.7032	0.7023
0.1253	14.0	9828	2.1676	0.7060	0.7082	0.7060	0.7049
0.0969	15.0	10530	2.1290	0.7103	0.7115	0.7103	0.7101

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.0+cu121
• Datasets 3.1.0
• Tokenizers 0.19.1