deberta_Energie

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: 0.0142
• Accuracy: 0.9913
• F1: 0.9913

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 32
• 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: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
1.4003	1.0	116	0.9262	0.6524	0.6025
0.7697	2.0	232	0.3836	0.8906	0.8899
0.3904	3.0	348	0.2468	0.9256	0.9191
0.2749	4.0	464	0.2202	0.9324	0.9283
0.2043	5.0	580	0.1122	0.9672	0.9673
0.1808	6.0	696	0.1004	0.9701	0.9706
0.1274	7.0	812	0.0822	0.9745	0.9747
0.1018	8.0	928	0.0673	0.9791	0.9794
0.0711	9.0	1044	0.0457	0.9870	0.9870
0.0609	10.0	1160	0.0370	0.9867	0.9867
0.0594	11.0	1276	0.0240	0.9886	0.9886
0.0332	12.0	1392	0.0182	0.9913	0.9913
0.0278	13.0	1508	0.0183	0.9908	0.9908
0.0281	14.0	1624	0.0142	0.9913	0.9913

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0