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: 0.8277
Accuracy: 0.895

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
0.3246	1.0	225	0.2699	0.9017
0.2407	2.0	450	0.2916	0.8967
0.1215	3.0	675	0.3070	0.9
0.0685	4.0	900	0.3909	0.885
0.0669	5.0	1125	0.4441	0.8783
0.0554	6.0	1350	0.5807	0.9017
0.0636	7.0	1575	0.6509	0.8967
0.1012	8.0	1800	0.5814	0.9017
0.0299	9.0	2025	0.6645	0.885
0.0537	10.0	2250	0.7621	0.8817
0.0367	11.0	2475	0.7288	0.8983
0.0022	12.0	2700	0.7692	0.8883
0.0254	13.0	2925	0.7258	0.9017
0.002	14.0	3150	0.7885	0.895
0.0002	15.0	3375	0.8594	0.8833
0.0001	16.0	3600	0.9131	0.8767
0.0002	17.0	3825	0.8422	0.895
0.0	18.0	4050	0.8469	0.89
0.0	19.0	4275	0.8455	0.895
0.0	20.0	4500	0.8023	0.8967
0.0	21.0	4725	0.7909	0.895
0.0	22.0	4950	0.8061	0.8933
0.0	23.0	5175	0.7917	0.8983
0.0	24.0	5400	0.8680	0.8917
0.0117	25.0	5625	0.8057	0.89
0.0	26.0	5850	0.7951	0.9
0.0	27.0	6075	0.7910	0.8983
0.0	28.0	6300	0.8032	0.8917
0.005	29.0	6525	0.7940	0.8933
0.0052	30.0	6750	0.8144	0.8933
0.0044	31.0	6975	0.8155	0.895
0.0	32.0	7200	0.8385	0.8983
0.0	33.0	7425	0.8031	0.8983
0.0	34.0	7650	0.7944	0.9
0.0	35.0	7875	0.8238	0.895
0.0	36.0	8100	0.8091	0.8933
0.0	37.0	8325	0.8071	0.8967
0.0	38.0	8550	0.8231	0.895
0.0	39.0	8775	0.8187	0.895
0.0	40.0	9000	0.8174	0.895
0.0	41.0	9225	0.8091	0.8967
0.0	42.0	9450	0.8169	0.895
0.0	43.0	9675	0.8228	0.895
0.0	44.0	9900	0.8259	0.8933
0.0	45.0	10125	0.8259	0.895
0.0	46.0	10350	0.8242	0.895
0.0026	47.0	10575	0.8264	0.895
0.0	48.0	10800	0.8265	0.895
0.0	49.0	11025	0.8271	0.895
0.0	50.0	11250	0.8277	0.895

Framework versions

Transformers 4.32.1
Pytorch 2.1.1+cu121
Datasets 2.12.0
Tokenizers 0.13.2

hkivancoral/smids_3x_deit_tiny_adamax_0001_fold3

smids_3x_deit_tiny_adamax_0001_fold3

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

hkivancoral/
smids_3x_deit_tiny_adamax_0001_fold3