smids_1x_deit_small_adamax_00001_fold1

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: 0.6979
• Accuracy: 0.8648

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
0.5894	1.0	76	0.5595	0.7830
0.4626	2.0	152	0.4242	0.8314
0.2923	3.0	228	0.3702	0.8531
0.2746	4.0	304	0.3590	0.8464
0.2111	5.0	380	0.3390	0.8631
0.1639	6.0	456	0.3346	0.8614
0.108	7.0	532	0.3491	0.8698
0.1272	8.0	608	0.3559	0.8681
0.132	9.0	684	0.3508	0.8648
0.0579	10.0	760	0.3614	0.8715
0.0748	11.0	836	0.3587	0.8681
0.0507	12.0	912	0.3817	0.8715
0.0272	13.0	988	0.4000	0.8781
0.0136	14.0	1064	0.4159	0.8731
0.0187	15.0	1140	0.4377	0.8648
0.0055	16.0	1216	0.4760	0.8698
0.0039	17.0	1292	0.5096	0.8648
0.0182	18.0	1368	0.5153	0.8698
0.0025	19.0	1444	0.5136	0.8698
0.0086	20.0	1520	0.5513	0.8664
0.0136	21.0	1596	0.5585	0.8765
0.0139	22.0	1672	0.5755	0.8648
0.0008	23.0	1748	0.5786	0.8765
0.0007	24.0	1824	0.6119	0.8748
0.013	25.0	1900	0.6090	0.8681
0.0086	26.0	1976	0.6766	0.8581
0.0066	27.0	2052	0.6214	0.8681
0.0074	28.0	2128	0.6445	0.8664
0.0116	29.0	2204	0.6664	0.8614
0.0003	30.0	2280	0.6454	0.8648
0.006	31.0	2356	0.6504	0.8681
0.0144	32.0	2432	0.6501	0.8698
0.0003	33.0	2508	0.6602	0.8698
0.0003	34.0	2584	0.6626	0.8648
0.0124	35.0	2660	0.6658	0.8698
0.0103	36.0	2736	0.6772	0.8698
0.0136	37.0	2812	0.6878	0.8648
0.0002	38.0	2888	0.6900	0.8681
0.0083	39.0	2964	0.6827	0.8631
0.0002	40.0	3040	0.6875	0.8698
0.0045	41.0	3116	0.6912	0.8664
0.0002	42.0	3192	0.6876	0.8614
0.0047	43.0	3268	0.6912	0.8631
0.0002	44.0	3344	0.7110	0.8598
0.0002	45.0	3420	0.6957	0.8648
0.0002	46.0	3496	0.6969	0.8648
0.0002	47.0	3572	0.6980	0.8648
0.0039	48.0	3648	0.6956	0.8681
0.0002	49.0	3724	0.6983	0.8648
0.0002	50.0	3800	0.6979	0.8648

Framework versions

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