learning_rate: 8e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH 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 Loss	Epoch	Step	Validation Loss	Accuracy
1.7425	1.0	225	1.5372	0.59
0.987	2.0	450	1.0627	0.69
0.5222	3.0	675	0.7102	0.73
0.4738	4.0	900	0.8310	0.74
0.2161	5.0	1125	0.7271	0.8
0.0103	6.0	1350	0.7824	0.84
0.063	7.0	1575	0.9816	0.79
0.012	8.0	1800	1.2869	0.8
0.0018	9.0	2025	0.9231	0.82
0.001	10.0	2250	0.9579	0.84
0.0008	11.0	2475	0.9185	0.84
0.0007	12.0	2700	1.0446	0.83
0.0007	13.0	2925	0.9800	0.83
0.0006	14.0	3150	1.0443	0.83
0.0006	15.0	3375	1.0320	0.83

ahachem140/distilhubert-finetuned-gtzan

distilhubert-finetuned-gtzan