BantuBERTa-vmw-finetuned-vmw-noaug

This model is a fine-tuned version of Kuongan/BantuBERTa-vmw-finetuned on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4673
• F1: 0.2169
• Roc Auc: 0.5714
• Accuracy: 0.4651

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: 32
• eval_batch_size: 32
• 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
0.1056	1.0	49	0.3296	0.1466	0.5417	0.4729
0.1022	2.0	98	0.3384	0.1506	0.5425	0.4806
0.0889	3.0	147	0.3528	0.1372	0.5439	0.4651
0.071	4.0	196	0.3808	0.1874	0.5581	0.4264
0.0659	5.0	245	0.3800	0.1664	0.5478	0.4690
0.0534	6.0	294	0.3999	0.1873	0.5564	0.4651
0.0407	7.0	343	0.4108	0.1439	0.5378	0.4535
0.0368	8.0	392	0.4235	0.1970	0.5590	0.4612
0.0247	9.0	441	0.4248	0.1724	0.5497	0.4729
0.0213	10.0	490	0.4313	0.2049	0.5648	0.4690
0.0164	11.0	539	0.4384	0.1918	0.5592	0.4690
0.0144	12.0	588	0.4576	0.2116	0.5694	0.4574
0.0138	13.0	637	0.4575	0.2075	0.5665	0.4729
0.0117	14.0	686	0.4694	0.2008	0.5638	0.4574
0.0114	15.0	735	0.4673	0.2169	0.5714	0.4651
0.0098	16.0	784	0.4703	0.2034	0.5646	0.4690
0.0096	17.0	833	0.4729	0.1943	0.5600	0.4574
0.0102	18.0	882	0.4707	0.1989	0.5619	0.4690
0.0101	19.0	931	0.4720	0.1990	0.5622	0.4690

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0