hushem_40x_deit_small_rms_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: 0.2918
• Accuracy: 0.9762

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
0.5087	1.0	219	0.4037	0.7619
0.1898	2.0	438	0.1339	0.9762
0.0553	3.0	657	0.0324	0.9762
0.0797	4.0	876	0.1848	0.9762
0.0341	5.0	1095	0.2228	0.9762
0.0296	6.0	1314	0.2257	0.9286
0.0744	7.0	1533	0.1717	0.9524
0.0049	8.0	1752	0.3696	0.9048
0.0089	9.0	1971	0.3392	0.9286
0.0001	10.0	2190	0.4146	0.9286
0.0322	11.0	2409	0.3832	0.9524
0.0165	12.0	2628	0.7717	0.9048
0.0	13.0	2847	0.2462	0.9762
0.0339	14.0	3066	0.0004	1.0
0.0335	15.0	3285	0.0062	1.0
0.0205	16.0	3504	0.2197	0.9524
0.0	17.0	3723	0.1117	0.9762
0.0	18.0	3942	0.1233	0.9762
0.0	19.0	4161	0.1357	0.9762
0.0	20.0	4380	0.1491	0.9762
0.0	21.0	4599	0.1602	0.9762
0.0	22.0	4818	0.1668	0.9762
0.0	23.0	5037	0.1701	0.9762
0.0	24.0	5256	0.1738	0.9762
0.0	25.0	5475	0.1788	0.9762
0.0	26.0	5694	0.1882	0.9762
0.0	27.0	5913	0.2002	0.9762
0.0	28.0	6132	0.2109	0.9762
0.0	29.0	6351	0.2232	0.9762
0.0	30.0	6570	0.2349	0.9762
0.0	31.0	6789	0.2441	0.9762
0.0	32.0	7008	0.2518	0.9762
0.0	33.0	7227	0.2582	0.9762
0.0	34.0	7446	0.2637	0.9762
0.0	35.0	7665	0.2684	0.9762
0.0	36.0	7884	0.2722	0.9762
0.0	37.0	8103	0.2755	0.9762
0.0	38.0	8322	0.2784	0.9762
0.0	39.0	8541	0.2809	0.9762
0.0	40.0	8760	0.2832	0.9762
0.0	41.0	8979	0.2850	0.9762
0.0	42.0	9198	0.2865	0.9762
0.0	43.0	9417	0.2879	0.9762
0.0	44.0	9636	0.2889	0.9762
0.0	45.0	9855	0.2898	0.9762
0.0	46.0	10074	0.2906	0.9762
0.0	47.0	10293	0.2911	0.9762
0.0	48.0	10512	0.2915	0.9762
0.0	49.0	10731	0.2917	0.9762
0.0	50.0	10950	0.2918	0.9762

Framework versions

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