iamaries/mdeberta_healthver

results

This model is a fine-tuned version of MoritzLaurer/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7 on the HealthVer dataset. It achieves the following results on the evaluation set:

• Loss: 0.6293
• Accuracy: 0.7681
• F1: 0.7721
• Precision: 0.7822
• Recall: 0.7681

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Use 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: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.6861	0.9980	248	0.6216	0.7392	0.7430	0.7566	0.7392
0.5114	2.0	497	0.5992	0.7646	0.7684	0.7769	0.7646
0.4074	2.9940	744	0.6293	0.7681	0.7721	0.7822	0.7681

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3