xlm-roberta-large-classifier

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

• Loss: 0.3918
• Accuracy: 0.8353
• F1: 0.7325

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 OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0	0	0.6059	0.7106	0.1957
No log	0.6006	188	0.4820	0.7826	0.6
No log	1.2013	376	0.4764	0.7858	0.5553
0.5275	1.8019	564	0.5046	0.7738	0.6519
0.5275	2.4026	752	0.4234	0.8233	0.7041
0.5275	3	939	0.3918	0.8353	0.7325

Framework versions

• Transformers 4.53.3
• Pytorch 2.6.0+cu124
• Datasets 4.0.0
• Tokenizers 0.21.2