tiny-audio-next-plus

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

• Loss: 0.3325

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: 0.001
• train_batch_size: 50
• eval_batch_size: 50
• seed: 43
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine_with_min_lr
• lr_scheduler_warmup_steps: 5000
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
0.3891	0.0470	2000	0.4868
0.3933	0.0939	4000	0.4988
0.3770	0.1409	6000	0.4885
0.3654	0.1879	8000	0.4802
0.3559	0.2348	10000	0.4621
0.3294	0.2818	12000	0.4519
0.3267	0.3287	14000	0.4353
0.3255	0.3757	16000	0.4314
0.3155	0.4227	18000	0.4185
0.3009	0.4696	20000	0.4137
0.2966	0.5166	22000	0.4031
0.3016	0.5636	24000	0.3930
0.2919	0.6105	26000	0.3963
0.2842	0.6575	28000	0.3966
0.2757	0.7045	30000	0.3915
0.2811	0.7514	32000	0.3801
0.2803	0.7984	34000	0.3814
0.2594	0.8453	36000	0.3727
0.2460	0.8923	38000	0.3671
0.2588	0.9393	40000	0.3607
0.2664	0.9862	42000	0.3618
0.2111	1.0332	44000	0.3581
0.2046	1.0802	46000	0.3639
0.2081	1.1271	48000	0.3586
0.2014	1.1741	50000	0.3615
0.2073	1.2211	52000	0.3573
0.2003	1.2680	54000	0.3545
0.2013	1.3150	56000	0.3547
0.1936	1.3619	58000	0.3559
0.1921	1.4089	60000	0.3469
0.1774	1.4559	62000	0.3473
0.2008	1.5028	64000	0.3444
0.1949	1.5498	66000	0.3459
0.1883	1.5968	68000	0.3462
0.1728	1.6437	70000	0.3425
0.1785	1.6907	72000	0.3413
0.1816	1.7377	74000	0.3396
0.1796	1.7846	76000	0.3359
0.1791	1.8316	78000	0.3358
0.2006	1.8786	80000	0.3367
0.1800	1.9255	82000	0.3317
0.1833	1.9725	84000	0.3344
0.1848	2.0	85172	0.3325

Framework versions

• Transformers 5.7.0
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.22.2