hushem_5x_deit_base_sgd_0001_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: 1.3736
• Accuracy: 0.3333

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
1.4044	1.0	27	1.3857	0.2889
1.4077	2.0	54	1.3850	0.2889
1.4004	3.0	81	1.3845	0.2667
1.4271	4.0	108	1.3839	0.2667
1.3916	5.0	135	1.3834	0.2889
1.3945	6.0	162	1.3828	0.2889
1.3833	7.0	189	1.3824	0.2889
1.3987	8.0	216	1.3819	0.2889
1.3885	9.0	243	1.3815	0.2889
1.4047	10.0	270	1.3810	0.2667
1.3682	11.0	297	1.3806	0.2889
1.3955	12.0	324	1.3802	0.2889
1.3888	13.0	351	1.3799	0.2889
1.3775	14.0	378	1.3795	0.2889
1.3829	15.0	405	1.3792	0.2889
1.3959	16.0	432	1.3788	0.2889
1.3739	17.0	459	1.3785	0.2889
1.3699	18.0	486	1.3782	0.2889
1.3964	19.0	513	1.3779	0.3111
1.3891	20.0	540	1.3776	0.3333
1.3791	21.0	567	1.3773	0.3333
1.3887	22.0	594	1.3770	0.3333
1.3832	23.0	621	1.3768	0.3333
1.3736	24.0	648	1.3765	0.3333
1.3564	25.0	675	1.3763	0.3111
1.3586	26.0	702	1.3761	0.3111
1.3706	27.0	729	1.3758	0.3111
1.3579	28.0	756	1.3756	0.3111
1.3705	29.0	783	1.3754	0.3111
1.3624	30.0	810	1.3752	0.3111
1.367	31.0	837	1.3751	0.3111
1.3534	32.0	864	1.3749	0.3333
1.3672	33.0	891	1.3747	0.3333
1.3654	34.0	918	1.3746	0.3333
1.3592	35.0	945	1.3744	0.3333
1.3511	36.0	972	1.3743	0.3333
1.3644	37.0	999	1.3742	0.3333
1.3508	38.0	1026	1.3741	0.3333
1.3516	39.0	1053	1.3740	0.3333
1.3605	40.0	1080	1.3739	0.3333
1.3566	41.0	1107	1.3739	0.3333
1.3647	42.0	1134	1.3738	0.3333
1.3671	43.0	1161	1.3737	0.3333
1.3471	44.0	1188	1.3737	0.3333
1.3589	45.0	1215	1.3737	0.3333
1.3648	46.0	1242	1.3736	0.3333
1.3521	47.0	1269	1.3736	0.3333
1.365	48.0	1296	1.3736	0.3333
1.3656	49.0	1323	1.3736	0.3333
1.3545	50.0	1350	1.3736	0.3333

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0