This model is a fine-tuned version of MIT/ast-finetuned-audioset-10-10-0.4593 on the GTZAN dataset. It achieves the following results on the evaluation set:

The following hyperparameters were used during training:

learning_rate: 0.0002
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
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_steps: 100
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7894	1.0	57	0.5304	0.86
0.5723	2.0	114	1.0449	0.69
0.5660	3.0	171	0.5600	0.82
0.3927	4.0	228	0.6672	0.83
0.2168	5.0	285	0.8590	0.8
0.1481	6.0	342	0.7175	0.84
0.0909	7.0	399	0.7492	0.83
0.0099	8.0	456	0.6389	0.87
0.0006	9.0	513	0.4595	0.88
0.0002	10.0	570	0.4532	0.89

vishal98m/ast-finetuned-audioset-10-10-0.4593-finetuned-gtzan

ast-finetuned-audioset-10-10-0.4593-finetuned-gtzan