This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 3.0098
Accuracy: 0.6444

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
0.3471	1.0	215	1.0795	0.5778
0.0972	2.0	430	0.8936	0.6889
0.0313	3.0	645	0.8957	0.6444
0.0058	4.0	860	1.0591	0.6889
0.0015	5.0	1075	1.2340	0.7111
0.0006	6.0	1290	1.2875	0.6889
0.0004	7.0	1505	1.3860	0.6889
0.0002	8.0	1720	1.4571	0.6889
0.0002	9.0	1935	1.5144	0.6667
0.0001	10.0	2150	1.5648	0.6889
0.0001	11.0	2365	1.6166	0.6667
0.0001	12.0	2580	1.6547	0.6889
0.0001	13.0	2795	1.7064	0.6667
0.0001	14.0	3010	1.7513	0.6667
0.0	15.0	3225	1.7849	0.6667
0.0	16.0	3440	1.8291	0.6667
0.0	17.0	3655	1.8746	0.6667
0.0	18.0	3870	1.9137	0.6667
0.0	19.0	4085	1.9589	0.6667
0.0	20.0	4300	2.0103	0.6667
0.0	21.0	4515	2.0484	0.6667
0.0	22.0	4730	2.0885	0.6667
0.0	23.0	4945	2.1272	0.6667
0.0	24.0	5160	2.1691	0.6667
0.0	25.0	5375	2.2032	0.6667
0.0	26.0	5590	2.2512	0.6667
0.0	27.0	5805	2.2928	0.6667
0.0	28.0	6020	2.3366	0.6667
0.0	29.0	6235	2.3684	0.6667
0.0	30.0	6450	2.4080	0.6667
0.0	31.0	6665	2.4434	0.6667
0.0	32.0	6880	2.4884	0.6667
0.0	33.0	7095	2.5184	0.6667
0.0	34.0	7310	2.5603	0.6667
0.0	35.0	7525	2.6005	0.6667
0.0	36.0	7740	2.6418	0.6444
0.0	37.0	7955	2.6720	0.6444
0.0	38.0	8170	2.7124	0.6444
0.0	39.0	8385	2.7569	0.6444
0.0	40.0	8600	2.7908	0.6444
0.0	41.0	8815	2.8243	0.6444
0.0	42.0	9030	2.8592	0.6444
0.0	43.0	9245	2.8889	0.6444
0.0	44.0	9460	2.9143	0.6444
0.0	45.0	9675	2.9439	0.6444
0.0	46.0	9890	2.9703	0.6444
0.0	47.0	10105	2.9822	0.6444
0.0	48.0	10320	3.0050	0.6444
0.0	49.0	10535	3.0086	0.6444
0.0	50.0	10750	3.0098	0.6444

Framework versions

Transformers 4.32.1
Pytorch 2.1.1+cu121
Datasets 2.12.0
Tokenizers 0.13.2

hkivancoral/hushem_40x_deit_tiny_adamax_00001_fold2

hushem_40x_deit_tiny_adamax_00001_fold2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

hkivancoral/
hushem_40x_deit_tiny_adamax_00001_fold2