Roberta_covidFact

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6179
• Accuracy: 0.694
• Auc: 0.498
• Precision: 0.0
• Recall: 0.0
• F1: 0.0
• F1-macro: 0.41
• F1-micro: 0.694
• F1-weighted: 0.569

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
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Auc	Precision	Recall	F1	F1-macro	F1-micro	F1-weighted
0.6882	0.5587	50	0.6186	0.694	0.504	0.0	0.0	0.0	0.41	0.694	0.569
0.633	1.1117	100	0.6167	0.694	0.524	0.0	0.0	0.0	0.41	0.694	0.569
0.6282	1.6704	150	0.6179	0.694	0.498	0.0	0.0	0.0	0.41	0.694	0.569

Framework versions

• Transformers 4.55.2
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.21.4