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: 0.9203
• Accuracy: 0.83

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.2556	1.0	113	2.1665	0.33
1.9061	2.0	226	1.7036	0.57
1.6449	3.0	339	1.3540	0.54
1.4074	4.0	452	1.1028	0.68
1.0121	5.0	565	0.9766	0.68
0.7831	6.0	678	0.8285	0.75
0.9829	7.0	791	0.7499	0.77
0.604	8.0	904	0.7093	0.77
0.6329	9.0	1017	0.7041	0.82
0.4323	10.0	1130	0.7244	0.83
0.1782	11.0	1243	0.7925	0.8
0.2571	12.0	1356	0.7031	0.84
0.1453	13.0	1469	0.7866	0.8
0.5919	14.0	1582	0.8135	0.82
0.307	15.0	1695	0.7489	0.85
0.2163	16.0	1808	0.9134	0.82
0.2081	17.0	1921	0.9109	0.85
0.1025	18.0	2034	0.9215	0.84
0.0415	19.0	2147	0.9542	0.84
0.481	20.0	2260	0.9203	0.83

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.2
• Tokenizers 0.13.3