smids_5x_deit_small_sgd_00001_fold2

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

• Loss: 0.9907
• Accuracy: 0.5108

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
1.0998	1.0	375	1.0701	0.4260
1.0872	2.0	750	1.0668	0.4276
1.0721	3.0	1125	1.0636	0.4343
1.068	4.0	1500	1.0604	0.4343
1.0595	5.0	1875	1.0573	0.4376
1.0368	6.0	2250	1.0542	0.4409
1.0324	7.0	2625	1.0513	0.4493
1.0454	8.0	3000	1.0484	0.4526
1.0424	9.0	3375	1.0456	0.4509
1.0256	10.0	3750	1.0428	0.4559
1.0778	11.0	4125	1.0401	0.4559
1.0323	12.0	4500	1.0375	0.4592
1.0063	13.0	4875	1.0349	0.4626
1.0266	14.0	5250	1.0324	0.4642
1.0153	15.0	5625	1.0299	0.4659
1.0	16.0	6000	1.0276	0.4692
1.0251	17.0	6375	1.0253	0.4692
1.0305	18.0	6750	1.0231	0.4725
1.0097	19.0	7125	1.0209	0.4725
1.02	20.0	7500	1.0189	0.4725
0.9981	21.0	7875	1.0168	0.4775
0.9952	22.0	8250	1.0149	0.4775
1.007	23.0	8625	1.0131	0.4859
1.0141	24.0	9000	1.0113	0.4875
1.0041	25.0	9375	1.0095	0.4875
1.0032	26.0	9750	1.0079	0.4875
1.0112	27.0	10125	1.0064	0.4875
0.9862	28.0	10500	1.0049	0.4892
0.9745	29.0	10875	1.0035	0.4892
0.9881	30.0	11250	1.0022	0.4925
0.9936	31.0	11625	1.0009	0.4908
0.9959	32.0	12000	0.9998	0.4892
0.9804	33.0	12375	0.9987	0.4875
0.9795	34.0	12750	0.9977	0.4958
0.9724	35.0	13125	0.9967	0.4975
0.9957	36.0	13500	0.9958	0.4992
0.9651	37.0	13875	0.9950	0.5025
0.9869	38.0	14250	0.9943	0.5042
0.9664	39.0	14625	0.9937	0.5075
0.9769	40.0	15000	0.9931	0.5092
0.9473	41.0	15375	0.9926	0.5108
0.9911	42.0	15750	0.9921	0.5108
0.9625	43.0	16125	0.9917	0.5108
0.9689	44.0	16500	0.9914	0.5108
0.9736	45.0	16875	0.9912	0.5108
0.9789	46.0	17250	0.9910	0.5108
0.9732	47.0	17625	0.9908	0.5108
0.9789	48.0	18000	0.9907	0.5108
0.987	49.0	18375	0.9907	0.5108
1.0128	50.0	18750	0.9907	0.5108

Framework versions

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