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: 1.5785
Accuracy: 0.2

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
No log	1.0	6	1.7159	0.2222
1.7117	2.0	12	1.7073	0.2222
1.7117	3.0	18	1.6994	0.2222
1.6883	4.0	24	1.6919	0.2222
1.6982	5.0	30	1.6843	0.2222
1.6982	6.0	36	1.6776	0.2222
1.6419	7.0	42	1.6712	0.2222
1.6419	8.0	48	1.6649	0.2222
1.6108	9.0	54	1.6588	0.2222
1.6345	10.0	60	1.6530	0.2222
1.6345	11.0	66	1.6473	0.2222
1.6108	12.0	72	1.6421	0.2222
1.6108	13.0	78	1.6369	0.2
1.6666	14.0	84	1.6323	0.2
1.6138	15.0	90	1.6282	0.2
1.6138	16.0	96	1.6241	0.2
1.5738	17.0	102	1.6203	0.2
1.5738	18.0	108	1.6167	0.2
1.5952	19.0	114	1.6131	0.2
1.555	20.0	120	1.6099	0.2
1.555	21.0	126	1.6070	0.2
1.5267	22.0	132	1.6040	0.2
1.5267	23.0	138	1.6012	0.2
1.5686	24.0	144	1.5988	0.2
1.5444	25.0	150	1.5966	0.2
1.5444	26.0	156	1.5944	0.2
1.544	27.0	162	1.5926	0.2
1.544	28.0	168	1.5907	0.2
1.5375	29.0	174	1.5889	0.2
1.5441	30.0	180	1.5873	0.2
1.5441	31.0	186	1.5857	0.2
1.5614	32.0	192	1.5845	0.2
1.5614	33.0	198	1.5832	0.2
1.5093	34.0	204	1.5821	0.2
1.5478	35.0	210	1.5812	0.2
1.5478	36.0	216	1.5804	0.2
1.5301	37.0	222	1.5798	0.2
1.5301	38.0	228	1.5793	0.2
1.4582	39.0	234	1.5789	0.2
1.5151	40.0	240	1.5786	0.2
1.5151	41.0	246	1.5785	0.2
1.5298	42.0	252	1.5785	0.2
1.5298	43.0	258	1.5785	0.2
1.548	44.0	264	1.5785	0.2
1.5172	45.0	270	1.5785	0.2
1.5172	46.0	276	1.5785	0.2
1.528	47.0	282	1.5785	0.2
1.528	48.0	288	1.5785	0.2
1.4968	49.0	294	1.5785	0.2
1.5413	50.0	300	1.5785	0.2

Framework versions

Transformers 4.35.0
Pytorch 2.1.0+cu118
Datasets 2.14.6
Tokenizers 0.14.1

hkivancoral/hushem_1x_deit_tiny_sgd_0001_fold2

hushem_1x_deit_tiny_sgd_0001_fold2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

hkivancoral/
hushem_1x_deit_tiny_sgd_0001_fold2