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: 0.6481
Accuracy: 0.7245

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
1.233	1.0	226	1.1760	0.3823
1.1228	2.0	452	1.1140	0.4023
1.086	3.0	678	1.0822	0.4174
1.0737	4.0	904	1.0561	0.4474
1.0492	5.0	1130	1.0319	0.4558
1.0539	6.0	1356	1.0095	0.4858
0.9458	7.0	1582	0.9879	0.5008
0.9572	8.0	1808	0.9681	0.5092
0.9484	9.0	2034	0.9485	0.5242
0.9168	10.0	2260	0.9300	0.5359
0.9317	11.0	2486	0.9125	0.5476
0.8523	12.0	2712	0.8952	0.5609
0.828	13.0	2938	0.8797	0.5843
0.8551	14.0	3164	0.8640	0.6010
0.8431	15.0	3390	0.8498	0.6127
0.7665	16.0	3616	0.8360	0.6177
0.7183	17.0	3842	0.8226	0.6210
0.7754	18.0	4068	0.8101	0.6344
0.7132	19.0	4294	0.7985	0.6394
0.7077	20.0	4520	0.7861	0.6544
0.6887	21.0	4746	0.7746	0.6628
0.7156	22.0	4972	0.7640	0.6661
0.7205	23.0	5198	0.7542	0.6761
0.6924	24.0	5424	0.7447	0.6811
0.668	25.0	5650	0.7359	0.6811
0.7303	26.0	5876	0.7276	0.6878
0.6039	27.0	6102	0.7198	0.6945
0.6316	28.0	6328	0.7126	0.6962
0.5808	29.0	6554	0.7060	0.6962
0.7521	30.0	6780	0.6997	0.7028
0.6067	31.0	7006	0.6939	0.7045
0.617	32.0	7232	0.6885	0.7062
0.5752	33.0	7458	0.6837	0.7062
0.5524	34.0	7684	0.6791	0.7078
0.645	35.0	7910	0.6750	0.7129
0.5855	36.0	8136	0.6712	0.7145
0.5981	37.0	8362	0.6677	0.7162
0.6026	38.0	8588	0.6646	0.7179
0.6372	39.0	8814	0.6617	0.7195
0.561	40.0	9040	0.6592	0.7179
0.5719	41.0	9266	0.6570	0.7179
0.5709	42.0	9492	0.6550	0.7195
0.6421	43.0	9718	0.6533	0.7212
0.5531	44.0	9944	0.6518	0.7245
0.6016	45.0	10170	0.6506	0.7245
0.6135	46.0	10396	0.6496	0.7245
0.5923	47.0	10622	0.6489	0.7245
0.5752	48.0	10848	0.6484	0.7245
0.5457	49.0	11074	0.6481	0.7245
0.586	50.0	11300	0.6481	0.7245

Framework versions

Transformers 4.32.1
Pytorch 2.1.1+cu121
Datasets 2.12.0
Tokenizers 0.13.2

hkivancoral/smids_3x_deit_tiny_sgd_0001_fold1

smids_3x_deit_tiny_sgd_0001_fold1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

hkivancoral/
smids_3x_deit_tiny_sgd_0001_fold1