results_soft_label

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

• Loss: 0.2341
• Mae: 0.0495
• Rmse: 0.1321
• Pearson Correlation: 0.9561
• Auc Roc: 0.9866
• Average Precision: 0.9803
• F1 At 0.5: 0.8754
• Precision At 0.5: 0.7871
• Recall At 0.5: 0.9862

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: 64
• seed: 42
• optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Mae	Rmse	Pearson Correlation	Auc Roc	Average Precision	F1 At 0.5	Precision At 0.5	Recall At 0.5
0.2484	1.0	11243	0.2585	0.0659	0.1608	0.9350	0.9802	0.9735	0.8617	0.7706	0.9772
0.2275	2.0	22486	0.2410	0.0541	0.1404	0.9506	0.9852	0.9800	0.8710	0.7811	0.9842
0.228	3.0	33729	0.2341	0.0495	0.1321	0.9561	0.9866	0.9803	0.8754	0.7871	0.9862

Framework versions

• Transformers 4.57.1
• Pytorch 2.9.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1