smids_5x_deit_base_adamax_0001_fold1

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: 0.6997
• Accuracy: 0.9182

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
0.2162	1.0	376	0.2261	0.9082
0.1041	2.0	752	0.2663	0.8965
0.0837	3.0	1128	0.3441	0.9015
0.0172	4.0	1504	0.4099	0.9048
0.0131	5.0	1880	0.4724	0.9048
0.0004	6.0	2256	0.4925	0.9065
0.0017	7.0	2632	0.6831	0.8965
0.0006	8.0	3008	0.5273	0.9015
0.0324	9.0	3384	0.5755	0.8998
0.0	10.0	3760	0.6569	0.9048
0.0009	11.0	4136	0.5873	0.9082
0.0003	12.0	4512	0.6069	0.9065
0.0	13.0	4888	0.5862	0.9082
0.0063	14.0	5264	0.6445	0.9048
0.0	15.0	5640	0.6277	0.9132
0.0	16.0	6016	0.7053	0.9032
0.0	17.0	6392	0.6033	0.9098
0.0	18.0	6768	0.6638	0.9065
0.0	19.0	7144	0.6432	0.9082
0.004	20.0	7520	0.6467	0.9115
0.0	21.0	7896	0.7009	0.9115
0.0	22.0	8272	0.7221	0.9048
0.0	23.0	8648	0.6516	0.9149
0.0	24.0	9024	0.6399	0.9149
0.0	25.0	9400	0.6382	0.9182
0.0034	26.0	9776	0.6520	0.9098
0.0	27.0	10152	0.6761	0.9115
0.0	28.0	10528	0.6436	0.9182
0.003	29.0	10904	0.6339	0.9115
0.0041	30.0	11280	0.6392	0.9132
0.0	31.0	11656	0.6548	0.9182
0.0	32.0	12032	0.6680	0.9149
0.0	33.0	12408	0.6562	0.9115
0.0	34.0	12784	0.6705	0.9165
0.0	35.0	13160	0.6801	0.9098
0.0	36.0	13536	0.6653	0.9182
0.0	37.0	13912	0.6565	0.9165
0.0	38.0	14288	0.6618	0.9215
0.0	39.0	14664	0.6597	0.9149
0.0	40.0	15040	0.6689	0.9165
0.0	41.0	15416	0.6826	0.9149
0.0	42.0	15792	0.6835	0.9132
0.0	43.0	16168	0.6862	0.9149
0.0	44.0	16544	0.6860	0.9182
0.0	45.0	16920	0.6904	0.9182
0.0027	46.0	17296	0.6967	0.9132
0.0	47.0	17672	0.6971	0.9182
0.0	48.0	18048	0.6989	0.9182
0.0	49.0	18424	0.7000	0.9182
0.0022	50.0	18800	0.6997	0.9182

Framework versions

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