smids_10x_deit_base_sgd_00001_fold2

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: 1.0091
• Accuracy: 0.5541

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
1.1301	1.0	750	1.1044	0.3361
1.0956	2.0	1500	1.1008	0.3394
1.0885	3.0	2250	1.0973	0.3527
1.0813	4.0	3000	1.0939	0.3677
1.0897	5.0	3750	1.0907	0.3794
1.0999	6.0	4500	1.0875	0.3794
1.0837	7.0	5250	1.0844	0.3810
1.0633	8.0	6000	1.0813	0.3860
1.0777	9.0	6750	1.0784	0.3977
1.0672	10.0	7500	1.0755	0.4027
1.0688	11.0	8250	1.0726	0.4143
1.0524	12.0	9000	1.0698	0.4193
1.0588	13.0	9750	1.0670	0.4276
1.0355	14.0	10500	1.0642	0.4443
1.0421	15.0	11250	1.0615	0.4526
1.0517	16.0	12000	1.0588	0.4526
1.0228	17.0	12750	1.0561	0.4609
1.052	18.0	13500	1.0535	0.4626
1.0453	19.0	14250	1.0509	0.4742
1.0307	20.0	15000	1.0483	0.4842
1.0308	21.0	15750	1.0459	0.4892
1.0369	22.0	16500	1.0434	0.5008
1.0173	23.0	17250	1.0411	0.5008
1.0178	24.0	18000	1.0388	0.5058
1.021	25.0	18750	1.0366	0.5075
1.0167	26.0	19500	1.0344	0.5075
1.0247	27.0	20250	1.0323	0.5125
1.0234	28.0	21000	1.0303	0.5208
1.003	29.0	21750	1.0283	0.5258
1.008	30.0	22500	1.0265	0.5308
1.0192	31.0	23250	1.0247	0.5341
1.0098	32.0	24000	1.0231	0.5374
0.9987	33.0	24750	1.0215	0.5408
0.9994	34.0	25500	1.0200	0.5408
1.0171	35.0	26250	1.0186	0.5408
1.013	36.0	27000	1.0173	0.5441
0.988	37.0	27750	1.0161	0.5441
0.9905	38.0	28500	1.0150	0.5491
0.9967	39.0	29250	1.0140	0.5491
0.9901	40.0	30000	1.0131	0.5557
0.9977	41.0	30750	1.0123	0.5557
1.0045	42.0	31500	1.0116	0.5557
0.9983	43.0	32250	1.0109	0.5541
0.9818	44.0	33000	1.0104	0.5541
0.9768	45.0	33750	1.0100	0.5541
0.9827	46.0	34500	1.0096	0.5541
0.9904	47.0	35250	1.0094	0.5541
0.9884	48.0	36000	1.0092	0.5541
0.9851	49.0	36750	1.0091	0.5541
0.9904	50.0	37500	1.0091	0.5541

Framework versions

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