ast-finetuned-audioset-16-16-0.442-finetuned-gtzan

This model is a fine-tuned version of ast-finetuned-audioset-16-16-0.442 on the GTZAN dataset. It achieves the following results on the evaluation set:

• Loss: 0.3315
• Accuracy: 0.93

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: 20
• eval_batch_size: 20
• seed: 42
• optimizer: Adam-8bits with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.8802	1.0	45	0.5267	0.85
0.3183	2.0	90	0.5893	0.81
0.1094	3.0	135	0.4421	0.89
0.0259	4.0	180	0.4100	0.88
0.0291	5.0	225	0.3695	0.9
0.0409	6.0	270	0.3071	0.91
0.0152	7.0	315	0.3482	0.92
0.0003	8.0	360	0.3187	0.94
0.0003	9.0	405	0.3258	0.93
0.0004	10.0	450	0.3315	0.93

Framework versions

• Transformers 4.40.1
• Pytorch 2.2.2+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1