smids_5x_beit_base_adamax_0001_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: 1.0655
• Accuracy: 0.8965

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
0.3493	1.0	376	0.3879	0.8581
0.2155	2.0	752	0.3019	0.8915
0.1776	3.0	1128	0.3875	0.8548
0.1242	4.0	1504	0.3809	0.8831
0.0896	5.0	1880	0.5028	0.8798
0.1253	6.0	2256	0.4979	0.8982
0.1104	7.0	2632	0.5865	0.8681
0.0316	8.0	3008	0.5613	0.8831
0.0721	9.0	3384	0.5293	0.8965
0.0201	10.0	3760	0.6272	0.8881
0.0359	11.0	4136	0.4934	0.8998
0.0744	12.0	4512	0.6114	0.8948
0.0347	13.0	4888	0.5456	0.9082
0.0311	14.0	5264	0.6131	0.8881
0.0168	15.0	5640	0.6543	0.8932
0.0168	16.0	6016	0.7183	0.8881
0.0016	17.0	6392	0.6732	0.8982
0.0267	18.0	6768	0.6217	0.9015
0.0052	19.0	7144	0.8606	0.8881
0.0397	20.0	7520	0.6236	0.8965
0.0267	21.0	7896	0.7627	0.8898
0.0186	22.0	8272	0.6922	0.8965
0.0249	23.0	8648	0.7332	0.8865
0.0032	24.0	9024	0.7665	0.8998
0.0275	25.0	9400	0.6785	0.8948
0.024	26.0	9776	0.7205	0.8915
0.0009	27.0	10152	0.7304	0.9015
0.0003	28.0	10528	0.7307	0.9065
0.0154	29.0	10904	0.7519	0.8965
0.0031	30.0	11280	0.8948	0.8932
0.0002	31.0	11656	0.8220	0.8998
0.0001	32.0	12032	0.7942	0.9048
0.0	33.0	12408	0.8498	0.9065
0.0055	34.0	12784	0.7753	0.8798
0.0001	35.0	13160	0.8717	0.8915
0.0	36.0	13536	0.9811	0.8865
0.0	37.0	13912	0.9556	0.8898
0.0003	38.0	14288	0.9804	0.8865
0.013	39.0	14664	0.9497	0.8965
0.0	40.0	15040	1.0094	0.8831
0.0	41.0	15416	0.9964	0.8881
0.0	42.0	15792	0.9367	0.8965
0.0	43.0	16168	1.0400	0.9015
0.0009	44.0	16544	1.0395	0.8948
0.0	45.0	16920	1.0420	0.8932
0.0031	46.0	17296	1.0873	0.8965
0.0	47.0	17672	1.0455	0.9032
0.0	48.0	18048	1.0612	0.8965
0.0	49.0	18424	1.0632	0.8998
0.0024	50.0	18800	1.0655	0.8965

Framework versions

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