smids_5x_beit_base_adamax_00001_fold1

This model is a fine-tuned version of microsoft/beit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.8312
• Accuracy: 0.9065

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
0.2349	1.0	376	0.2964	0.8848
0.2022	2.0	752	0.2944	0.8932
0.1706	3.0	1128	0.2893	0.8965
0.0767	4.0	1504	0.3105	0.9015
0.0646	5.0	1880	0.3471	0.9015
0.0505	6.0	2256	0.3777	0.9015
0.0505	7.0	2632	0.4146	0.9115
0.0821	8.0	3008	0.4739	0.9115
0.0331	9.0	3384	0.5133	0.9082
0.0097	10.0	3760	0.5125	0.9065
0.0368	11.0	4136	0.5327	0.9098
0.0236	12.0	4512	0.6377	0.8881
0.0306	13.0	4888	0.6671	0.9015
0.0605	14.0	5264	0.6154	0.9048
0.0306	15.0	5640	0.6497	0.9082
0.0004	16.0	6016	0.6905	0.9098
0.0062	17.0	6392	0.7456	0.9082
0.0157	18.0	6768	0.7362	0.9048
0.0117	19.0	7144	0.8082	0.8965
0.0001	20.0	7520	0.7613	0.9098
0.0049	21.0	7896	0.7376	0.9115
0.0013	22.0	8272	0.7490	0.9098
0.0339	23.0	8648	0.7577	0.9132
0.0009	24.0	9024	0.7847	0.9098
0.0161	25.0	9400	0.7983	0.9098
0.0079	26.0	9776	0.7734	0.8948
0.0004	27.0	10152	0.7368	0.9015
0.0005	28.0	10528	0.7478	0.9098
0.0059	29.0	10904	0.7755	0.9065
0.0012	30.0	11280	0.8338	0.9082
0.0142	31.0	11656	0.7783	0.9115
0.0002	32.0	12032	0.7615	0.9165
0.0004	33.0	12408	0.7711	0.9098
0.0127	34.0	12784	0.7865	0.9165
0.0032	35.0	13160	0.8207	0.9132
0.0006	36.0	13536	0.8174	0.9098
0.0001	37.0	13912	0.7992	0.9165
0.0	38.0	14288	0.8040	0.9082
0.0001	39.0	14664	0.8011	0.9132
0.0005	40.0	15040	0.8052	0.9115
0.0001	41.0	15416	0.8158	0.9082
0.0001	42.0	15792	0.8157	0.9098
0.0	43.0	16168	0.8347	0.9065
0.0004	44.0	16544	0.8096	0.9048
0.0087	45.0	16920	0.8231	0.9065
0.0003	46.0	17296	0.8362	0.9065
0.0002	47.0	17672	0.8291	0.9098
0.0046	48.0	18048	0.8341	0.9082
0.0134	49.0	18424	0.8309	0.9065
0.0004	50.0	18800	0.8312	0.9065

Framework versions

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