distilhubert-finetuned-gtzan

This model is a fine-tuned version of ntu-spml/distilhubert on the GTZAN dataset. It achieves the following results on the evaluation set:

• Loss: 0.5819
• Accuracy: 0.89

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.1087	1.0	113	2.0088	0.52
1.3935	2.0	226	1.3173	0.65
0.8685	3.0	339	1.0099	0.74
0.7505	4.0	452	0.7838	0.76
0.5099	5.0	565	0.6810	0.83
0.4902	6.0	678	0.5436	0.86
0.2815	7.0	791	0.5144	0.84
0.2062	8.0	904	0.5575	0.84
0.0847	9.0	1017	0.5333	0.88
0.0254	10.0	1130	0.5955	0.86
0.0245	11.0	1243	0.5819	0.89
0.0282	12.0	1356	0.6459	0.86
0.0119	13.0	1469	0.6356	0.89
0.0089	14.0	1582	0.6942	0.87
0.0085	15.0	1695	0.6543	0.88

Framework versions

• Transformers 4.57.6
• Pytorch 2.8.0+cu128
• Datasets 4.8.4
• Tokenizers 0.22.2