smids_10x_deit_base_sgd_0001_fold1

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

• Loss: 0.4637
• Accuracy: 0.8197

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
1.0699	1.0	751	1.0865	0.3372
1.0294	2.0	1502	1.0518	0.4207
0.9671	3.0	2253	1.0151	0.4992
0.8907	4.0	3004	0.9742	0.5526
0.8635	5.0	3755	0.9306	0.6043
0.8115	6.0	4506	0.8888	0.6461
0.7615	7.0	5257	0.8493	0.6694
0.7205	8.0	6008	0.8123	0.6861
0.6698	9.0	6759	0.7780	0.7078
0.6144	10.0	7510	0.7466	0.7195
0.6353	11.0	8261	0.7180	0.7295
0.5516	12.0	9012	0.6921	0.7396
0.5555	13.0	9763	0.6685	0.7446
0.5272	14.0	10514	0.6476	0.7546
0.5021	15.0	11265	0.6291	0.7596
0.4905	16.0	12016	0.6125	0.7713
0.4771	17.0	12767	0.5975	0.7696
0.4486	18.0	13518	0.5841	0.7730
0.4932	19.0	14269	0.5723	0.7713
0.4372	20.0	15020	0.5615	0.7746
0.3961	21.0	15771	0.5521	0.7780
0.455	22.0	16522	0.5434	0.7796
0.4028	23.0	17273	0.5355	0.7896
0.4361	24.0	18024	0.5285	0.7913
0.4568	25.0	18775	0.5220	0.7930
0.5129	26.0	19526	0.5160	0.7947
0.4225	27.0	20277	0.5106	0.7997
0.4409	28.0	21028	0.5056	0.7997
0.4065	29.0	21779	0.5010	0.7997
0.4201	30.0	22530	0.4969	0.8013
0.3612	31.0	23281	0.4930	0.8047
0.5003	32.0	24032	0.4895	0.8063
0.4064	33.0	24783	0.4864	0.8080
0.4	34.0	25534	0.4834	0.8097
0.4159	35.0	26285	0.4807	0.8097
0.4048	36.0	27036	0.4783	0.8130
0.4636	37.0	27787	0.4761	0.8147
0.3893	38.0	28538	0.4740	0.8147
0.3469	39.0	29289	0.4723	0.8147
0.4394	40.0	30040	0.4706	0.8164
0.4132	41.0	30791	0.4692	0.8180
0.3775	42.0	31542	0.4680	0.8197
0.3957	43.0	32293	0.4669	0.8197
0.4064	44.0	33044	0.4660	0.8197
0.4124	45.0	33795	0.4652	0.8197
0.3879	46.0	34546	0.4646	0.8197
0.3807	47.0	35297	0.4642	0.8197
0.4327	48.0	36048	0.4639	0.8197
0.3727	49.0	36799	0.4638	0.8197
0.3716	50.0	37550	0.4637	0.8197

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2