hushem_40x_deit_base_sgd_00001_fold4

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: 1.3940
• Accuracy: 0.3095

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
1.4075	1.0	219	1.4139	0.2857
1.4006	2.0	438	1.4128	0.2857
1.397	3.0	657	1.4119	0.2857
1.4009	4.0	876	1.4109	0.2857
1.4192	5.0	1095	1.4100	0.2857
1.4068	6.0	1314	1.4091	0.2857
1.4024	7.0	1533	1.4083	0.2857
1.3965	8.0	1752	1.4075	0.2857
1.3783	9.0	1971	1.4067	0.3095
1.3738	10.0	2190	1.4060	0.3095
1.3936	11.0	2409	1.4053	0.3095
1.3746	12.0	2628	1.4046	0.3095
1.3536	13.0	2847	1.4040	0.3095
1.4005	14.0	3066	1.4033	0.3095
1.3798	15.0	3285	1.4027	0.3095
1.3748	16.0	3504	1.4022	0.3095
1.3581	17.0	3723	1.4016	0.3095
1.3695	18.0	3942	1.4011	0.3095
1.366	19.0	4161	1.4006	0.3095
1.3735	20.0	4380	1.4001	0.3095
1.3732	21.0	4599	1.3997	0.3095
1.3632	22.0	4818	1.3992	0.3095
1.3525	23.0	5037	1.3988	0.3095
1.3845	24.0	5256	1.3984	0.3095
1.363	25.0	5475	1.3980	0.3095
1.3693	26.0	5694	1.3977	0.3095
1.3693	27.0	5913	1.3973	0.3095
1.3914	28.0	6132	1.3970	0.3095
1.3857	29.0	6351	1.3967	0.3095
1.3681	30.0	6570	1.3964	0.3095
1.3619	31.0	6789	1.3962	0.3095
1.3666	32.0	7008	1.3959	0.3095
1.3733	33.0	7227	1.3957	0.3095
1.3572	34.0	7446	1.3955	0.3095
1.3715	35.0	7665	1.3953	0.3095
1.3581	36.0	7884	1.3951	0.3095
1.3453	37.0	8103	1.3949	0.3095
1.3666	38.0	8322	1.3948	0.3095
1.3416	39.0	8541	1.3946	0.3095
1.3435	40.0	8760	1.3945	0.3095
1.3731	41.0	8979	1.3944	0.3095
1.3652	42.0	9198	1.3943	0.3095
1.3499	43.0	9417	1.3942	0.3095
1.3629	44.0	9636	1.3941	0.3095
1.3332	45.0	9855	1.3941	0.3095
1.3535	46.0	10074	1.3940	0.3095
1.3876	47.0	10293	1.3940	0.3095
1.363	48.0	10512	1.3940	0.3095
1.3575	49.0	10731	1.3940	0.3095
1.3466	50.0	10950	1.3940	0.3095

Framework versions

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