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: 1.2548
Accuracy: 0.5111

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
1.7252	1.0	215	1.5329	0.2667
1.5501	2.0	430	1.4609	0.3111
1.4663	3.0	645	1.4250	0.3111
1.4117	4.0	860	1.4032	0.2889
1.3533	5.0	1075	1.3870	0.2667
1.3221	6.0	1290	1.3733	0.2889
1.3111	7.0	1505	1.3613	0.3111
1.2698	8.0	1720	1.3509	0.3111
1.2425	9.0	1935	1.3418	0.3111
1.2243	10.0	2150	1.3338	0.3778
1.2016	11.0	2365	1.3268	0.3778
1.1128	12.0	2580	1.3203	0.3556
1.174	13.0	2795	1.3136	0.3556
1.1731	14.0	3010	1.3081	0.4
1.141	15.0	3225	1.3031	0.4
1.1163	16.0	3440	1.2979	0.4
1.1128	17.0	3655	1.2946	0.4222
1.0806	18.0	3870	1.2916	0.4222
1.0332	19.0	4085	1.2893	0.3778
1.0358	20.0	4300	1.2875	0.4
1.0352	21.0	4515	1.2855	0.4
1.0257	22.0	4730	1.2838	0.4
1.0362	23.0	4945	1.2822	0.4
1.0137	24.0	5160	1.2805	0.4
1.0067	25.0	5375	1.2787	0.4222
0.9834	26.0	5590	1.2771	0.4667
0.9889	27.0	5805	1.2753	0.4667
0.9291	28.0	6020	1.2744	0.4667
0.9563	29.0	6235	1.2728	0.4667
0.9949	30.0	6450	1.2710	0.4667
0.9331	31.0	6665	1.2698	0.4667
0.9189	32.0	6880	1.2683	0.4889
0.8977	33.0	7095	1.2667	0.4889
0.9506	34.0	7310	1.2657	0.4889
0.9018	35.0	7525	1.2644	0.4889
0.9085	36.0	7740	1.2632	0.4889
0.9525	37.0	7955	1.2617	0.4889
0.9147	38.0	8170	1.2608	0.4889
0.8837	39.0	8385	1.2597	0.5111
0.9228	40.0	8600	1.2588	0.5111
0.8773	41.0	8815	1.2582	0.5111
0.8964	42.0	9030	1.2574	0.5111
0.8892	43.0	9245	1.2568	0.5111
0.8986	44.0	9460	1.2562	0.5111
0.9114	45.0	9675	1.2557	0.5111
0.8745	46.0	9890	1.2553	0.5111
0.9224	47.0	10105	1.2551	0.5111
0.9229	48.0	10320	1.2549	0.5111
0.9087	49.0	10535	1.2549	0.5111
0.9371	50.0	10750	1.2548	0.5111

Framework versions

Transformers 4.32.1
Pytorch 2.1.0+cu121
Datasets 2.12.0
Tokenizers 0.13.2

hkivancoral/hushem_40x_deit_small_sgd_0001_fold2

hushem_40x_deit_small_sgd_0001_fold2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

hkivancoral/
hushem_40x_deit_small_sgd_0001_fold2