roberta-large-binary-classification

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

• Loss: 0.6983
• Accuracy: 0.7580
• F1 Macro: 0.7453
• Precision Macro: 0.7498
• Recall Macro: 0.7425
• Auc: 0.7941

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: 16
• eval_batch_size: 16
• seed: 42
• 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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro	Precision Macro	Recall Macro	Auc
No log	1.0	79	0.6751	0.5955	0.3733	0.2978	0.5	0.6194
No log	2.0	158	0.6642	0.5955	0.3733	0.2978	0.5	0.6210
No log	3.0	237	0.5609	0.7102	0.6895	0.7003	0.6859	0.7701
No log	4.0	316	0.5676	0.7070	0.6907	0.6954	0.6883	0.7734
No log	5.0	395	0.6983	0.7580	0.7453	0.7498	0.7425	0.7941
No log	6.0	474	0.7766	0.7420	0.7319	0.7322	0.7316	0.7802
0.4887	7.0	553	1.1879	0.7452	0.7266	0.7399	0.7217	0.7761
0.4887	8.0	632	1.6676	0.7484	0.7242	0.7504	0.7180	0.7789
0.4887	9.0	711	1.6440	0.7548	0.7364	0.7511	0.7310	0.7889
0.4887	10.0	790	1.7092	0.7548	0.7364	0.7511	0.7310	0.7928

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1