smids_5x_deit_base_sgd_00001_fold5

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: 1.0685
• Accuracy: 0.4283

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
1.1186	1.0	375	1.1098	0.3483
1.1102	2.0	750	1.1079	0.3417
1.1136	3.0	1125	1.1060	0.345
1.1131	4.0	1500	1.1043	0.35
1.0975	5.0	1875	1.1026	0.3433
1.1051	6.0	2250	1.1010	0.345
1.0892	7.0	2625	1.0994	0.3417
1.0802	8.0	3000	1.0978	0.3533
1.0951	9.0	3375	1.0964	0.3517
1.0929	10.0	3750	1.0949	0.3517
1.0628	11.0	4125	1.0935	0.3533
1.0809	12.0	4500	1.0922	0.36
1.0566	13.0	4875	1.0909	0.375
1.0849	14.0	5250	1.0897	0.38
1.0684	15.0	5625	1.0884	0.3817
1.0868	16.0	6000	1.0873	0.3817
1.0653	17.0	6375	1.0861	0.3817
1.0768	18.0	6750	1.0850	0.385
1.0758	19.0	7125	1.0839	0.385
1.0932	20.0	7500	1.0829	0.3883
1.072	21.0	7875	1.0819	0.39
1.06	22.0	8250	1.0809	0.3917
1.0521	23.0	8625	1.0800	0.3967
1.0558	24.0	9000	1.0792	0.3983
1.0773	25.0	9375	1.0783	0.4
1.0609	26.0	9750	1.0775	0.4
1.0495	27.0	10125	1.0767	0.4017
1.0658	28.0	10500	1.0760	0.4
1.0475	29.0	10875	1.0753	0.4083
1.0538	30.0	11250	1.0746	0.415
1.0455	31.0	11625	1.0740	0.4133
1.0741	32.0	12000	1.0734	0.415
1.0518	33.0	12375	1.0728	0.4167
1.04	34.0	12750	1.0723	0.4183
1.0566	35.0	13125	1.0718	0.4167
1.0416	36.0	13500	1.0714	0.4167
1.0546	37.0	13875	1.0709	0.4217
1.0514	38.0	14250	1.0706	0.4233
1.0481	39.0	14625	1.0702	0.425
1.0581	40.0	15000	1.0699	0.4267
1.0484	41.0	15375	1.0696	0.4267
1.0544	42.0	15750	1.0694	0.4267
1.0499	43.0	16125	1.0691	0.4267
1.0424	44.0	16500	1.0690	0.4267
1.0515	45.0	16875	1.0688	0.4283
1.0389	46.0	17250	1.0687	0.4283
1.0556	47.0	17625	1.0686	0.4283
1.0595	48.0	18000	1.0685	0.4283
1.0528	49.0	18375	1.0685	0.4283
1.0533	50.0	18750	1.0685	0.4283

Framework versions

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