smids_10x_deit_small_adamax_001_fold3

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: 1.1653
• Accuracy: 0.8983

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: 0.001
• 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
0.2967	1.0	750	0.3026	0.8833
0.2176	2.0	1500	0.3634	0.865
0.1482	3.0	2250	0.3577	0.8917
0.1878	4.0	3000	0.3366	0.89
0.2072	5.0	3750	0.4027	0.8783
0.1159	6.0	4500	0.4215	0.9017
0.1092	7.0	5250	0.4522	0.875
0.1055	8.0	6000	0.5168	0.8733
0.0404	9.0	6750	0.4834	0.8867
0.0353	10.0	7500	0.5506	0.895
0.0316	11.0	8250	0.5770	0.88
0.0204	12.0	9000	0.7013	0.88
0.0769	13.0	9750	0.7049	0.8833
0.0408	14.0	10500	0.5508	0.8983
0.014	15.0	11250	0.6644	0.8883
0.0112	16.0	12000	0.7305	0.895
0.0118	17.0	12750	0.6466	0.8967
0.0015	18.0	13500	0.7382	0.89
0.0022	19.0	14250	0.9099	0.8967
0.0028	20.0	15000	0.8123	0.8883
0.0003	21.0	15750	0.7936	0.895
0.0021	22.0	16500	0.8670	0.89
0.0001	23.0	17250	0.8387	0.89
0.0001	24.0	18000	0.9036	0.8867
0.0071	25.0	18750	0.9933	0.8967
0.0003	26.0	19500	0.9103	0.8933
0.0	27.0	20250	0.9486	0.8983
0.0	28.0	21000	0.9480	0.89
0.0	29.0	21750	1.0149	0.8983
0.0	30.0	22500	0.9710	0.89
0.0	31.0	23250	0.8903	0.9017
0.0	32.0	24000	0.9900	0.8983
0.0	33.0	24750	0.9812	0.8967
0.0	34.0	25500	1.0802	0.8917
0.0	35.0	26250	1.0127	0.9
0.0	36.0	27000	1.0499	0.8917
0.0	37.0	27750	1.0711	0.895
0.0018	38.0	28500	1.1040	0.8983
0.0	39.0	29250	1.0513	0.9017
0.0	40.0	30000	1.1398	0.9
0.0	41.0	30750	1.1537	0.9
0.0	42.0	31500	1.1196	0.9
0.0	43.0	32250	1.1395	0.8967
0.0	44.0	33000	1.1136	0.9017
0.0	45.0	33750	1.1523	0.895
0.0	46.0	34500	1.1446	0.9
0.0	47.0	35250	1.1542	0.8967
0.0	48.0	36000	1.1560	0.8983
0.0	49.0	36750	1.1589	0.8983
0.0	50.0	37500	1.1653	0.8983

Framework versions

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