nicolauduran45/deberta-v3-base4scientific-claim-verification

classifier

This model is a fine-tuned version of microsoft/deberta-v3-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9760
• F1-micro: 0.7354
• F1-macro: 0.7202
• F1-weighted: 0.7314
• Accuracy: 0.7354

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: 8
• seed: 42
• 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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	F1-micro	F1-macro	F1-weighted	Accuracy
No log	1.0	386	0.7548	0.6615	0.6527	0.6627	0.6615
0.8384	2.0	772	0.6518	0.7134	0.6981	0.7073	0.7134
0.6001	3.0	1158	0.6752	0.7419	0.7313	0.7408	0.7419
0.3926	4.0	1544	0.7489	0.7354	0.7232	0.7304	0.7354
0.3926	5.0	1930	0.9760	0.7354	0.7202	0.7314	0.7354

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1