roberta-base_binary

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

• Loss: 0.1729
• Precision: 0.8178
• Recall: 0.6136
• F1: 0.7012
• F0.5: 0.7668
• Macro Precision: 0.8824
• Macro Recall: 0.7971
• Macro F1: 0.8323
• Macro F0.5: 0.8602

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: 3e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• 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: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	F0.5	Macro Precision	Macro Recall	Macro F1	Macro F0.5
0.1963	1.0	1926	0.1702	0.8148	0.6179	0.7028	0.7660	0.8814	0.7991	0.8333	0.8601
0.1621	1.9992	3850	0.1698	0.8027	0.6472	0.7166	0.7659	0.8772	0.8124	0.8405	0.8613

Framework versions

• Transformers 4.50.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1