distilhubert-finetuned-gtzan-finetuned-gtzan

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

• Loss: 0.9179
• Accuracy: 0.84

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.15
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.3025	1.0	57	2.2928	0.2
2.2609	2.0	114	2.2683	0.27
1.9479	3.0	171	1.9099	0.4
1.211	4.0	228	1.5258	0.39
0.9834	5.0	285	1.4254	0.52
0.6456	6.0	342	1.3216	0.57
0.5043	7.0	399	1.1890	0.69
0.4696	8.0	456	1.0764	0.8
0.3204	9.0	513	0.9564	0.82
0.3164	10.0	570	0.9101	0.83
0.2334	11.0	627	0.9021	0.84
0.217	12.0	684	0.9051	0.84
0.1781	13.0	741	0.9118	0.84
0.1203	14.0	798	0.9153	0.85
0.0639	15.0	855	0.9179	0.84

Framework versions

• Transformers 4.36.0.dev0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1