learning_rate: 0.0008
train_batch_size: 256
eval_batch_size: 256
seed: 42
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
num_epochs: 5
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.4609	0.2844	5000	1.3989	0.6532
1.3211	0.5689	10000	1.2739	0.6803
1.2531	0.8533	15000	1.2132	0.6942
1.1875	1.1377	20000	1.1762	0.7041
1.157	1.4222	25000	1.1460	0.7111
1.144	1.7066	30000	1.1184	0.7163
1.1217	1.9910	35000	1.0880	0.7247
1.0831	2.2754	40000	1.0729	0.7280
1.0761	2.5599	45000	1.0593	0.7312
1.0565	2.8443	50000	1.0480	0.7346
1.0149	3.1287	55000	1.0356	0.7380
1.0102	3.4132	60000	1.0263	0.7401
1.0014	3.6976	65000	1.0122	0.7437
0.9972	3.9820	70000	1.0028	0.7459
0.9556	4.2665	75000	0.9971	0.7474
0.9606	4.5509	80000	0.9904	0.7496
0.9544	4.8353	85000	0.9842	0.7507

VinayHajare/quickdraw-mobilevit-small

results