deberta_v3_MP_ft

This model is a fine-tuned version of microsoft/deberta-v3-large on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0218
• Macro F1: 0.0068
• Micro F1: 0.0236
• Macro Precision: 0.1547
• Macro Recall: 0.0037
• Micro Precision: 0.8725
• Micro Recall: 0.0120
• Exact Match Ratio: 0.0164
• Macro Roc Auc: 0.7865

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: cosine
• lr_scheduler_warmup_steps: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Macro F1	Micro F1	Macro Precision	Macro Recall	Micro Precision	Micro Recall	Exact Match Ratio	Macro Roc Auc
0.0287	6.5789	500	0.0227	0.0025	0.0084	0.1303	0.0013	0.8873	0.0042	0.0059	0.7649
0.0277	10.0	760	0.0218	0.0062	0.0220	0.1611	0.0033	0.8872	0.0111	0.0058	0.7843

Framework versions

• Transformers 5.12.1
• Pytorch 2.5.1+cu121
• Datasets 5.0.1
• Tokenizers 0.22.2