smids_5x_deit_small_sgd_00001_fold3

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: 1.0078
• Accuracy: 0.4883

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
1.0625	1.0	375	1.0854	0.38
1.055	2.0	750	1.0820	0.3817
1.0441	3.0	1125	1.0787	0.3817
1.0543	4.0	1500	1.0755	0.3833
1.0717	5.0	1875	1.0724	0.3833
1.0405	6.0	2250	1.0694	0.3833
1.0573	7.0	2625	1.0664	0.3867
1.052	8.0	3000	1.0635	0.3933
1.0402	9.0	3375	1.0606	0.395
1.026	10.0	3750	1.0579	0.3967
1.0363	11.0	4125	1.0552	0.4017
1.044	12.0	4500	1.0526	0.4033
1.0227	13.0	4875	1.0501	0.4117
1.0237	14.0	5250	1.0477	0.4133
1.0137	15.0	5625	1.0453	0.4183
1.005	16.0	6000	1.0431	0.4167
1.0298	17.0	6375	1.0409	0.4167
1.0209	18.0	6750	1.0387	0.4183
1.0296	19.0	7125	1.0366	0.425
1.0081	20.0	7500	1.0346	0.4283
0.9849	21.0	7875	1.0327	0.4317
1.0033	22.0	8250	1.0308	0.44
1.0003	23.0	8625	1.0290	0.4417
1.0236	24.0	9000	1.0274	0.445
0.9768	25.0	9375	1.0257	0.4533
0.9963	26.0	9750	1.0242	0.4567
0.9973	27.0	10125	1.0227	0.46
1.025	28.0	10500	1.0213	0.4617
0.9786	29.0	10875	1.0199	0.465
1.0006	30.0	11250	1.0187	0.4667
1.0183	31.0	11625	1.0175	0.47
0.9871	32.0	12000	1.0164	0.4733
0.9751	33.0	12375	1.0154	0.4733
0.9558	34.0	12750	1.0144	0.475
0.9521	35.0	13125	1.0135	0.475
0.975	36.0	13500	1.0127	0.475
0.9912	37.0	13875	1.0119	0.4783
0.9818	38.0	14250	1.0112	0.48
0.9973	39.0	14625	1.0106	0.4817
0.9737	40.0	15000	1.0101	0.4833
0.9571	41.0	15375	1.0096	0.4833
0.9497	42.0	15750	1.0092	0.4833
0.9898	43.0	16125	1.0088	0.485
0.9733	44.0	16500	1.0085	0.485
0.9695	45.0	16875	1.0083	0.4833
0.9603	46.0	17250	1.0081	0.4867
0.9924	47.0	17625	1.0079	0.4867
0.9781	48.0	18000	1.0079	0.4867
1.0064	49.0	18375	1.0078	0.4883
0.9488	50.0	18750	1.0078	0.4883

Framework versions

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