smids_3x_deit_base_sgd_00001_fold3

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: 1.0784
• Accuracy: 0.4267

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
1.1104	1.0	225	1.1064	0.345
1.1269	2.0	450	1.1051	0.355
1.1082	3.0	675	1.1039	0.3667
1.0954	4.0	900	1.1026	0.365
1.1069	5.0	1125	1.1015	0.36
1.1012	6.0	1350	1.1003	0.3667
1.1071	7.0	1575	1.0992	0.3733
1.1242	8.0	1800	1.0982	0.3767
1.086	9.0	2025	1.0971	0.3783
1.0866	10.0	2250	1.0961	0.3867
1.0948	11.0	2475	1.0952	0.3833
1.0863	12.0	2700	1.0942	0.385
1.0844	13.0	2925	1.0933	0.3833
1.0933	14.0	3150	1.0925	0.3867
1.0947	15.0	3375	1.0916	0.3917
1.084	16.0	3600	1.0908	0.39
1.0986	17.0	3825	1.0900	0.395
1.0824	18.0	4050	1.0893	0.3967
1.0832	19.0	4275	1.0886	0.395
1.0894	20.0	4500	1.0879	0.3967
1.0841	21.0	4725	1.0872	0.4
1.0872	22.0	4950	1.0865	0.405
1.0916	23.0	5175	1.0859	0.4117
1.0847	24.0	5400	1.0853	0.4117
1.0901	25.0	5625	1.0848	0.41
1.0732	26.0	5850	1.0842	0.41
1.0848	27.0	6075	1.0837	0.4133
1.0818	28.0	6300	1.0832	0.415
1.0774	29.0	6525	1.0828	0.415
1.0812	30.0	6750	1.0823	0.4183
1.0886	31.0	6975	1.0819	0.4183
1.0712	32.0	7200	1.0815	0.42
1.0744	33.0	7425	1.0812	0.42
1.0756	34.0	7650	1.0808	0.425
1.0664	35.0	7875	1.0805	0.4267
1.0977	36.0	8100	1.0802	0.4283
1.0683	37.0	8325	1.0799	0.4283
1.0735	38.0	8550	1.0797	0.4267
1.0832	39.0	8775	1.0795	0.4267
1.0815	40.0	9000	1.0793	0.4267
1.0823	41.0	9225	1.0791	0.425
1.0956	42.0	9450	1.0789	0.425
1.0851	43.0	9675	1.0788	0.425
1.0774	44.0	9900	1.0787	0.4267
1.0466	45.0	10125	1.0786	0.4267
1.0871	46.0	10350	1.0785	0.4267
1.0722	47.0	10575	1.0784	0.4267
1.069	48.0	10800	1.0784	0.4267
1.0654	49.0	11025	1.0784	0.4267
1.0659	50.0	11250	1.0784	0.4267

Framework versions

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