UIT-deberta-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: 0.4347
• F1: 0.7574
• Roc Auc: 0.8115
• Accuracy: 0.5036

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.5508	1.0	139	0.4985	0.1682	0.5518	0.1643
0.4245	2.0	278	0.4152	0.6348	0.7259	0.3881
0.3521	3.0	417	0.3579	0.7279	0.7918	0.4747
0.2247	4.0	556	0.3752	0.7206	0.7825	0.4675
0.2045	5.0	695	0.3588	0.7561	0.8109	0.5181
0.1466	6.0	834	0.4355	0.7353	0.7928	0.4765
0.0911	7.0	973	0.4347	0.7574	0.8115	0.5036
0.0731	8.0	1112	0.4792	0.7233	0.7869	0.4639
0.0576	9.0	1251	0.5001	0.7505	0.8056	0.5054
0.0376	10.0	1390	0.5599	0.7404	0.7973	0.4928
0.0248	11.0	1529	0.5667	0.7416	0.7982	0.4964
0.0147	12.0	1668	0.5936	0.7442	0.8033	0.4982
0.0148	13.0	1807	0.5873	0.7531	0.8133	0.5
0.009	14.0	1946	0.6078	0.7465	0.8029	0.5072
0.0122	15.0	2085	0.6217	0.7475	0.8045	0.4946
0.0057	16.0	2224	0.6321	0.7479	0.8037	0.5072
0.0072	17.0	2363	0.6359	0.7451	0.8021	0.5036
0.0061	18.0	2502	0.6397	0.7480	0.8044	0.5018
0.0063	19.0	2641	0.6387	0.7466	0.8031	0.5
0.0063	20.0	2780	0.6394	0.7470	0.8034	0.5018

Framework versions

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