smids_3x_deit_base_adamax_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: 0.6129
• Accuracy: 0.9183

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
0.3685	1.0	225	0.3281	0.8817
0.2394	2.0	450	0.2668	0.9033
0.1551	3.0	675	0.2550	0.9167
0.1239	4.0	900	0.2574	0.9117
0.1058	5.0	1125	0.2783	0.915
0.058	6.0	1350	0.2899	0.91
0.0386	7.0	1575	0.3339	0.9033
0.0293	8.0	1800	0.3749	0.9117
0.0298	9.0	2025	0.3833	0.9167
0.013	10.0	2250	0.4190	0.91
0.0241	11.0	2475	0.4405	0.91
0.0009	12.0	2700	0.4672	0.9017
0.0005	13.0	2925	0.4613	0.9083
0.0003	14.0	3150	0.4827	0.9133
0.0004	15.0	3375	0.4942	0.915
0.001	16.0	3600	0.5082	0.915
0.0002	17.0	3825	0.5550	0.9117
0.0001	18.0	4050	0.5345	0.915
0.0001	19.0	4275	0.5425	0.915
0.0001	20.0	4500	0.5481	0.9167
0.0001	21.0	4725	0.5684	0.9167
0.0001	22.0	4950	0.5664	0.9117
0.0	23.0	5175	0.5625	0.915
0.0001	24.0	5400	0.5934	0.9133
0.0037	25.0	5625	0.6215	0.9067
0.0001	26.0	5850	0.5781	0.9183
0.0	27.0	6075	0.5846	0.9183
0.0	28.0	6300	0.5796	0.9183
0.0021	29.0	6525	0.6062	0.9167
0.0066	30.0	6750	0.5817	0.9167
0.0028	31.0	6975	0.6008	0.915
0.0	32.0	7200	0.5847	0.9167
0.0	33.0	7425	0.5946	0.9183
0.0	34.0	7650	0.6041	0.9133
0.0	35.0	7875	0.5926	0.915
0.0	36.0	8100	0.5946	0.92
0.0	37.0	8325	0.6079	0.915
0.0	38.0	8550	0.5973	0.9217
0.0	39.0	8775	0.5993	0.9183
0.0	40.0	9000	0.6003	0.9183
0.0	41.0	9225	0.6120	0.915
0.0	42.0	9450	0.6070	0.92
0.0	43.0	9675	0.6080	0.9183
0.0	44.0	9900	0.6062	0.9217
0.0	45.0	10125	0.6095	0.9183
0.0	46.0	10350	0.6121	0.9183
0.0019	47.0	10575	0.6121	0.9183
0.0	48.0	10800	0.6134	0.9183
0.0	49.0	11025	0.6133	0.9183
0.0	50.0	11250	0.6129	0.9183

Framework versions

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