smids_1x_deit_tiny_sgd_00001_fold5

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: 1.1969
• Accuracy: 0.3683

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
1.2818	1.0	75	1.3526	0.35
1.2881	2.0	150	1.3438	0.35
1.2902	3.0	225	1.3353	0.3533
1.3491	4.0	300	1.3273	0.3517
1.2508	5.0	375	1.3195	0.355
1.2901	6.0	450	1.3122	0.355
1.2792	7.0	525	1.3053	0.3583
1.2973	8.0	600	1.2988	0.3583
1.3051	9.0	675	1.2924	0.3583
1.3668	10.0	750	1.2863	0.3583
1.2982	11.0	825	1.2805	0.3633
1.1991	12.0	900	1.2750	0.3617
1.2833	13.0	975	1.2699	0.3617
1.2768	14.0	1050	1.2648	0.36
1.2691	15.0	1125	1.2602	0.36
1.2029	16.0	1200	1.2557	0.3617
1.2189	17.0	1275	1.2513	0.3667
1.2814	18.0	1350	1.2472	0.3683
1.1777	19.0	1425	1.2435	0.37
1.2006	20.0	1500	1.2398	0.3683
1.3016	21.0	1575	1.2363	0.3717
1.2664	22.0	1650	1.2331	0.3683
1.1963	23.0	1725	1.2301	0.37
1.2239	24.0	1800	1.2272	0.37
1.1881	25.0	1875	1.2244	0.37
1.2397	26.0	1950	1.2219	0.3717
1.1817	27.0	2025	1.2194	0.3717
1.2303	28.0	2100	1.2172	0.3733
1.253	29.0	2175	1.2151	0.3733
1.1936	30.0	2250	1.2131	0.3733
1.2173	31.0	2325	1.2113	0.3733
1.153	32.0	2400	1.2096	0.3733
1.2175	33.0	2475	1.2080	0.3733
1.2243	34.0	2550	1.2065	0.3733
1.1302	35.0	2625	1.2052	0.3717
1.1855	36.0	2700	1.2040	0.37
1.1832	37.0	2775	1.2029	0.37
1.1866	38.0	2850	1.2019	0.365
1.2112	39.0	2925	1.2010	0.365
1.199	40.0	3000	1.2002	0.3667
1.1826	41.0	3075	1.1995	0.3667
1.2211	42.0	3150	1.1988	0.3683
1.2093	43.0	3225	1.1983	0.3683
1.2039	44.0	3300	1.1979	0.3683
1.1848	45.0	3375	1.1975	0.3683
1.2445	46.0	3450	1.1973	0.3683
1.1786	47.0	3525	1.1971	0.3683
1.1742	48.0	3600	1.1970	0.3683
1.1159	49.0	3675	1.1969	0.3683
1.1807	50.0	3750	1.1969	0.3683

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0