smids_1x_deit_small_adamax_0001_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: 0.5996
• Accuracy: 0.9083

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: 0.0001
• 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
0.3937	1.0	75	0.3211	0.8633
0.2826	2.0	150	0.3394	0.8733
0.1652	3.0	225	0.2814	0.8967
0.141	4.0	300	0.2984	0.9
0.0291	5.0	375	0.3773	0.9017
0.0153	6.0	450	0.4868	0.8733
0.0215	7.0	525	0.4578	0.9067
0.0108	8.0	600	0.5844	0.875
0.0251	9.0	675	0.5045	0.9033
0.0276	10.0	750	0.7184	0.875
0.0071	11.0	825	0.5704	0.89
0.0105	12.0	900	0.5523	0.9067
0.0005	13.0	975	0.5683	0.9033
0.0002	14.0	1050	0.5274	0.9033
0.0001	15.0	1125	0.5432	0.895
0.0001	16.0	1200	0.5742	0.8983
0.0001	17.0	1275	0.5512	0.9067
0.0071	18.0	1350	0.5548	0.9033
0.0043	19.0	1425	0.5622	0.9083
0.0063	20.0	1500	0.5939	0.9033
0.0	21.0	1575	0.5379	0.9133
0.0	22.0	1650	0.5428	0.91
0.0037	23.0	1725	0.5469	0.9067
0.0	24.0	1800	0.5517	0.9083
0.0	25.0	1875	0.5493	0.9083
0.0032	26.0	1950	0.5544	0.915
0.0	27.0	2025	0.5586	0.9117
0.0	28.0	2100	0.5623	0.9117
0.0	29.0	2175	0.5631	0.9117
0.0	30.0	2250	0.5628	0.91
0.0	31.0	2325	0.5710	0.9117
0.0	32.0	2400	0.5769	0.91
0.0	33.0	2475	0.5763	0.91
0.0048	34.0	2550	0.5811	0.9117
0.0073	35.0	2625	0.5738	0.9117
0.0031	36.0	2700	0.5751	0.91
0.0023	37.0	2775	0.5897	0.9133
0.0	38.0	2850	0.5810	0.91
0.0	39.0	2925	0.5835	0.91
0.0	40.0	3000	0.5848	0.9083
0.0	41.0	3075	0.5898	0.91
0.0027	42.0	3150	0.5994	0.915
0.0028	43.0	3225	0.5922	0.91
0.0	44.0	3300	0.5948	0.91
0.0027	45.0	3375	0.5970	0.91
0.0025	46.0	3450	0.5958	0.9083
0.0051	47.0	3525	0.6023	0.9133
0.0	48.0	3600	0.5981	0.9083
0.0	49.0	3675	0.6006	0.91
0.0045	50.0	3750	0.5996	0.9083

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0