UIT-NO-PREdeberta-v3-base-finetuned

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: 0.7415
• F1: 0.7580
• Roc Auc: 0.8223
• Accuracy: 0.4964

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 adamw_torch 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: 100
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.5738	1.0	139	0.5473	0.1947	0.5461	0.1751
0.4173	2.0	278	0.4194	0.5997	0.7189	0.3736
0.3402	3.0	417	0.3829	0.6887	0.7809	0.4314
0.1937	4.0	556	0.3563	0.7213	0.7914	0.4747
0.1756	5.0	695	0.3888	0.7271	0.7883	0.4856
0.1307	6.0	834	0.4043	0.7351	0.7959	0.4964
0.0963	7.0	973	0.4520	0.7434	0.8055	0.4982
0.0645	8.0	1112	0.4992	0.7259	0.7935	0.4892
0.0396	9.0	1251	0.5118	0.7570	0.8209	0.4964
0.0298	10.0	1390	0.5700	0.7516	0.8107	0.4892
0.0403	11.0	1529	0.6019	0.7440	0.8071	0.4892
0.014	12.0	1668	0.6476	0.7355	0.8008	0.4964
0.0095	13.0	1807	0.6871	0.7364	0.8007	0.4477
0.0127	14.0	1946	0.6843	0.7489	0.8097	0.4928
0.0098	15.0	2085	0.7105	0.7504	0.8082	0.4928
0.0039	16.0	2224	0.7187	0.7512	0.8119	0.4928
0.0044	17.0	2363	0.7415	0.7580	0.8223	0.4964
0.0035	18.0	2502	0.7614	0.7419	0.8051	0.4874
0.0033	19.0	2641	0.7527	0.7460	0.8137	0.4874
0.0027	20.0	2780	0.7640	0.7491	0.8113	0.5036
0.0014	21.0	2919	0.7682	0.7490	0.8113	0.4964
0.0014	22.0	3058	0.7804	0.7514	0.8124	0.5018
0.002	23.0	3197	0.7642	0.7567	0.8197	0.5072
0.0018	24.0	3336	0.7774	0.7537	0.8159	0.5018
0.0016	25.0	3475	0.7758	0.7545	0.8165	0.5054
0.0016	26.0	3614	0.7811	0.7580	0.8195	0.5108
0.0017	27.0	3753	0.7819	0.7558	0.8167	0.5108
0.0015	28.0	3892	0.7833	0.7574	0.8180	0.5126
0.0014	29.0	4031	0.7844	0.7565	0.8169	0.5126
0.001	30.0	4170	0.7846	0.7562	0.8165	0.5108

Framework versions

• Transformers 4.48.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.21.0