smids_5x_beit_base_adamax_0001_fold5

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: 1.0130
• Accuracy: 0.9067

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
0.2609	1.0	375	0.4392	0.83
0.2435	2.0	750	0.3135	0.87
0.1555	3.0	1125	0.3719	0.885
0.1553	4.0	1500	0.4465	0.89
0.1	5.0	1875	0.4697	0.88
0.0953	6.0	2250	0.5131	0.895
0.1058	7.0	2625	0.5042	0.9033
0.0095	8.0	3000	0.6171	0.9033
0.0607	9.0	3375	0.5787	0.89
0.0225	10.0	3750	0.6004	0.8983
0.0522	11.0	4125	0.6713	0.89
0.0354	12.0	4500	0.8122	0.9
0.0281	13.0	4875	0.7547	0.8833
0.0538	14.0	5250	0.6866	0.88
0.0498	15.0	5625	0.7040	0.8783
0.0034	16.0	6000	0.6946	0.8883
0.0375	17.0	6375	0.7067	0.88
0.0372	18.0	6750	0.8461	0.875
0.0081	19.0	7125	0.5733	0.9
0.025	20.0	7500	0.6029	0.9167
0.0105	21.0	7875	0.6183	0.885
0.0342	22.0	8250	0.5174	0.9217
0.0291	23.0	8625	0.5708	0.9083
0.0316	24.0	9000	0.7866	0.8833
0.0002	25.0	9375	0.8031	0.895
0.0548	26.0	9750	0.7954	0.8933
0.0233	27.0	10125	0.8188	0.895
0.0003	28.0	10500	0.7997	0.9
0.0063	29.0	10875	0.8708	0.89
0.0025	30.0	11250	0.8386	0.8967
0.0008	31.0	11625	0.8998	0.8833
0.0	32.0	12000	0.9085	0.8967
0.0005	33.0	12375	0.7875	0.905
0.0	34.0	12750	0.9329	0.8983
0.0001	35.0	13125	0.7985	0.9017
0.0	36.0	13500	0.8234	0.8983
0.0	37.0	13875	0.8947	0.9033
0.005	38.0	14250	0.9096	0.9067
0.0291	39.0	14625	0.9293	0.9117
0.0006	40.0	15000	0.8881	0.9117
0.0	41.0	15375	1.0854	0.8967
0.0003	42.0	15750	0.9486	0.8983
0.0	43.0	16125	0.9324	0.91
0.0	44.0	16500	0.9408	0.9083
0.0	45.0	16875	1.0069	0.9067
0.0	46.0	17250	1.0803	0.9
0.013	47.0	17625	1.0261	0.905
0.0	48.0	18000	1.0163	0.9067
0.0	49.0	18375	1.0208	0.9083
0.0021	50.0	18750	1.0130	0.9067

Framework versions

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