learning_rate: 3e-05
train_batch_size: 16
eval_batch_size: 16
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: cosine_with_restarts
lr_scheduler_warmup_ratio: 0.15
num_epochs: 30

Training Loss	Epoch	Step	Accuracy	Validation Loss
2.2385	1.0	50	0.34	2.2256
1.7883	2.0	100	0.545	1.7487
1.4788	3.0	150	0.68	1.4187
1.1262	4.0	200	0.69	1.1004
0.8664	5.0	250	0.735	0.9532
0.7772	6.0	300	0.745	0.8106
0.4455	7.0	350	0.81	0.7057
0.3719	8.0	400	0.815	0.6467
0.3716	9.0	450	0.805	0.6164
0.223	10.0	500	0.825	0.5887
0.1382	11.0	550	0.83	0.5941
0.0729	12.0	600	0.84	0.5911
0.0518	13.0	650	0.84	0.6116
0.0388	14.0	700	0.835	0.6217
0.0304	15.0	750	0.84	0.6340
0.0266	16.0	800	0.84	0.6407
0.026	17.0	850	0.85	0.6428
0.0238	18.0	900	0.84	0.6457
0.0244	19.0	950	0.85	0.6457
0.0278	20.0	1000	0.845	0.6466
0.0676	19.0	1026	0.6300	0.8533
0.0271	20.0	1080	0.6714	0.8467
0.0165	21.0	1134	0.6385	0.8533
0.0158	22.0	1188	0.6895	0.8667
0.0292	23.0	1242	0.6982	0.86
0.0232	24.0	1296	0.6870	0.86
0.0099	25.0	1350	0.6774	0.8667
0.0104	26.0	1404	0.6821	0.86
0.0101	27.0	1458	0.6773	0.86
0.01	28.0	1512	0.6790	0.86
0.0097	29.0	1566	0.6779	0.86
0.0093	30.0	1620	0.6759	0.86

sinha-mayank-900/distilhubert-finetuned-gtzan

distilhubert-finetuned-gtzan