smids_5x_deit_tiny_adamax_0001_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: 0.7374
• Accuracy: 0.9115

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
0.2308	1.0	376	0.2388	0.9115
0.1317	2.0	752	0.2582	0.9098
0.1686	3.0	1128	0.4270	0.8831
0.0405	4.0	1504	0.4962	0.8781
0.0366	5.0	1880	0.4884	0.8965
0.0181	6.0	2256	0.6032	0.8982
0.0062	7.0	2632	0.6900	0.8965
0.0021	8.0	3008	0.5787	0.9115
0.0022	9.0	3384	0.7270	0.8932
0.0002	10.0	3760	0.6268	0.9115
0.0012	11.0	4136	0.5883	0.9082
0.0001	12.0	4512	0.5732	0.9165
0.0	13.0	4888	0.5583	0.9215
0.011	14.0	5264	0.7089	0.9032
0.0002	15.0	5640	0.7954	0.8815
0.0	16.0	6016	0.6772	0.8965
0.0	17.0	6392	0.6923	0.9032
0.0	18.0	6768	0.7006	0.8982
0.0	19.0	7144	0.6930	0.9032
0.0149	20.0	7520	0.7767	0.8948
0.0	21.0	7896	0.6643	0.9132
0.0	22.0	8272	0.6887	0.9015
0.0	23.0	8648	0.6854	0.8998
0.0001	24.0	9024	0.6666	0.9132
0.0	25.0	9400	0.6680	0.9082
0.0033	26.0	9776	0.6701	0.9115
0.0	27.0	10152	0.6769	0.8998
0.0	28.0	10528	0.6638	0.9115
0.0042	29.0	10904	0.6671	0.9165
0.004	30.0	11280	0.6642	0.9115
0.0	31.0	11656	0.6776	0.9098
0.0	32.0	12032	0.6835	0.9098
0.0	33.0	12408	0.7324	0.9048
0.0	34.0	12784	0.7298	0.9048
0.0	35.0	13160	0.7063	0.9098
0.0	36.0	13536	0.7145	0.9098
0.0	37.0	13912	0.7213	0.9082
0.0	38.0	14288	0.7239	0.9098
0.0	39.0	14664	0.7296	0.9065
0.0	40.0	15040	0.7280	0.9082
0.0	41.0	15416	0.7172	0.9098
0.0	42.0	15792	0.7224	0.9098
0.0	43.0	16168	0.7245	0.9115
0.0	44.0	16544	0.7307	0.9098
0.0	45.0	16920	0.7306	0.9098
0.0026	46.0	17296	0.7273	0.9098
0.0	47.0	17672	0.7340	0.9132
0.0	48.0	18048	0.7346	0.9132
0.0	49.0	18424	0.7359	0.9132
0.0022	50.0	18800	0.7374	0.9115

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.1+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2