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: 0.3138
Accuracy: 0.8537

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: 0.001
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
1.2269	1.0	220	1.3889	0.3171
0.9985	2.0	440	1.1915	0.3902
0.8593	3.0	660	1.0700	0.5366
0.7571	4.0	880	0.9669	0.5854
0.6777	5.0	1100	0.8606	0.6341
0.5274	6.0	1320	0.7627	0.6585
0.4414	7.0	1540	0.6648	0.7317
0.4175	8.0	1760	0.6018	0.7805
0.316	9.0	1980	0.5359	0.7805
0.2588	10.0	2200	0.4774	0.8049
0.2327	11.0	2420	0.4274	0.8537
0.1962	12.0	2640	0.3959	0.8293
0.1844	13.0	2860	0.3504	0.8293
0.1541	14.0	3080	0.3380	0.8293
0.1252	15.0	3300	0.3134	0.8293
0.1083	16.0	3520	0.2915	0.8537
0.0934	17.0	3740	0.2801	0.8537
0.092	18.0	3960	0.2741	0.8780
0.083	19.0	4180	0.2729	0.8780
0.0654	20.0	4400	0.2600	0.8780
0.0561	21.0	4620	0.2597	0.8780
0.0451	22.0	4840	0.2669	0.8780
0.0472	23.0	5060	0.2513	0.8780
0.0354	24.0	5280	0.2691	0.8780
0.0367	25.0	5500	0.2575	0.8780
0.0263	26.0	5720	0.2498	0.8780
0.0333	27.0	5940	0.2592	0.8780
0.0309	28.0	6160	0.2598	0.8780
0.0217	29.0	6380	0.2806	0.8780
0.0156	30.0	6600	0.2646	0.8780
0.0138	31.0	6820	0.2633	0.8780
0.0242	32.0	7040	0.2880	0.8780
0.0128	33.0	7260	0.2733	0.8780
0.0109	34.0	7480	0.2902	0.8780
0.011	35.0	7700	0.2855	0.8780
0.0099	36.0	7920	0.3131	0.8537
0.0084	37.0	8140	0.2898	0.8780
0.0211	38.0	8360	0.3199	0.8537
0.014	39.0	8580	0.3071	0.8537
0.0076	40.0	8800	0.2872	0.8780
0.0086	41.0	9020	0.3095	0.8537
0.0068	42.0	9240	0.3068	0.8537
0.0085	43.0	9460	0.3042	0.8537
0.0079	44.0	9680	0.3170	0.8537
0.0089	45.0	9900	0.3144	0.8537
0.005	46.0	10120	0.3132	0.8537
0.0073	47.0	10340	0.3173	0.8537
0.01	48.0	10560	0.3150	0.8537
0.0068	49.0	10780	0.3144	0.8537
0.0068	50.0	11000	0.3138	0.8537

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_001_fold5

hushem_40x_deit_small_sgd_001_fold5

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

hkivancoral/
hushem_40x_deit_small_sgd_001_fold5