smids_5x_deit_small_adamax_0001_fold4

This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.2568
• Accuracy: 0.895

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.1785	1.0	750	0.3310	0.8867
0.1427	2.0	1500	0.4997	0.86
0.0558	3.0	2250	0.6477	0.8833
0.0755	4.0	3000	0.8076	0.8783
0.0696	5.0	3750	0.8523	0.885
0.0129	6.0	4500	0.8649	0.8917
0.0009	7.0	5250	0.8612	0.895
0.0	8.0	6000	0.9953	0.8883
0.0001	9.0	6750	0.9803	0.89
0.0001	10.0	7500	0.9507	0.895
0.0	11.0	8250	1.0047	0.8983
0.0	12.0	9000	1.0208	0.885
0.0	13.0	9750	1.0442	0.8867
0.0	14.0	10500	0.9977	0.89
0.0	15.0	11250	1.0546	0.8917
0.0087	16.0	12000	1.1978	0.885
0.0001	17.0	12750	1.0539	0.9017
0.0	18.0	13500	1.1390	0.8917
0.0	19.0	14250	1.0555	0.9
0.0	20.0	15000	1.0783	0.8983
0.0	21.0	15750	1.1342	0.89
0.0	22.0	16500	1.1482	0.895
0.0	23.0	17250	1.1356	0.8933
0.0	24.0	18000	1.0819	0.9
0.0	25.0	18750	1.0556	0.8967
0.0116	26.0	19500	1.1710	0.8917
0.0	27.0	20250	1.1214	0.8967
0.0	28.0	21000	1.1327	0.8967
0.0	29.0	21750	1.1390	0.895
0.0	30.0	22500	1.1576	0.8967
0.0	31.0	23250	1.1495	0.8933
0.0	32.0	24000	1.1623	0.9
0.0	33.0	24750	1.1633	0.895
0.0	34.0	25500	1.1868	0.895
0.0	35.0	26250	1.1906	0.8983
0.0	36.0	27000	1.2000	0.8967
0.0	37.0	27750	1.2102	0.8983
0.0	38.0	28500	1.2162	0.8967
0.0	39.0	29250	1.2243	0.895
0.0	40.0	30000	1.2297	0.895
0.0	41.0	30750	1.2339	0.8933
0.0	42.0	31500	1.2401	0.8933
0.0	43.0	32250	1.2422	0.8933
0.0	44.0	33000	1.2459	0.8933
0.0	45.0	33750	1.2496	0.895
0.0	46.0	34500	1.2523	0.895
0.0	47.0	35250	1.2541	0.895
0.0	48.0	36000	1.2558	0.895
0.0	49.0	36750	1.2566	0.895
0.0	50.0	37500	1.2568	0.895

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2