smids_3x_deit_base_adamax_00001_fold1

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

• Loss: 0.6128
• Accuracy: 0.8948

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: 1e-05
• 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
0.3987	1.0	226	0.3422	0.8831
0.2561	2.0	452	0.2788	0.8998
0.1593	3.0	678	0.2675	0.8998
0.1432	4.0	904	0.2725	0.8881
0.1177	5.0	1130	0.2842	0.8982
0.0548	6.0	1356	0.2961	0.9048
0.017	7.0	1582	0.3174	0.9032
0.0253	8.0	1808	0.3381	0.9032
0.0176	9.0	2034	0.3719	0.8998
0.0045	10.0	2260	0.4333	0.8948
0.0047	11.0	2486	0.4348	0.8932
0.0045	12.0	2712	0.4668	0.8982
0.0004	13.0	2938	0.4893	0.8998
0.0211	14.0	3164	0.4896	0.8948
0.0003	15.0	3390	0.5202	0.8932
0.0002	16.0	3616	0.5276	0.8881
0.0192	17.0	3842	0.5347	0.8932
0.0001	18.0	4068	0.5437	0.8982
0.0215	19.0	4294	0.5418	0.8915
0.0001	20.0	4520	0.5545	0.8932
0.0001	21.0	4746	0.5523	0.8948
0.0193	22.0	4972	0.5629	0.8915
0.0179	23.0	5198	0.5596	0.8948
0.0001	24.0	5424	0.5887	0.8965
0.0001	25.0	5650	0.5699	0.8982
0.0001	26.0	5876	0.5700	0.8948
0.0	27.0	6102	0.5958	0.8932
0.0001	28.0	6328	0.5917	0.8932
0.0	29.0	6554	0.5908	0.8982
0.0001	30.0	6780	0.5873	0.8965
0.0001	31.0	7006	0.5806	0.8982
0.0	32.0	7232	0.6122	0.8932
0.0043	33.0	7458	0.6075	0.8932
0.0	34.0	7684	0.5998	0.8982
0.0	35.0	7910	0.5938	0.8948
0.0	36.0	8136	0.5898	0.8932
0.0001	37.0	8362	0.5968	0.8948
0.0	38.0	8588	0.6080	0.8982
0.0	39.0	8814	0.6052	0.8948
0.0	40.0	9040	0.6086	0.8965
0.0027	41.0	9266	0.6076	0.8965
0.002	42.0	9492	0.6009	0.8965
0.0	43.0	9718	0.6115	0.8948
0.0	44.0	9944	0.6097	0.8965
0.0	45.0	10170	0.6098	0.8932
0.0	46.0	10396	0.6112	0.8948
0.0	47.0	10622	0.6121	0.8948
0.0	48.0	10848	0.6134	0.8948
0.0	49.0	11074	0.6129	0.8948
0.0	50.0	11300	0.6128	0.8948

Framework versions

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