smids_5x_beit_base_sgd_00001_fold4

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: 1.1429
• Accuracy: 0.3883

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
1.2481	1.0	375	1.3387	0.3367
1.2853	2.0	750	1.3297	0.34
1.2245	3.0	1125	1.3209	0.34
1.2269	4.0	1500	1.3124	0.34
1.1881	5.0	1875	1.3044	0.34
1.1962	6.0	2250	1.2965	0.3417
1.1766	7.0	2625	1.2890	0.345
1.1422	8.0	3000	1.2816	0.3467
1.1265	9.0	3375	1.2745	0.3483
1.1532	10.0	3750	1.2676	0.3517
1.1456	11.0	4125	1.2609	0.3533
1.1221	12.0	4500	1.2545	0.355
1.1397	13.0	4875	1.2483	0.355
1.1323	14.0	5250	1.2422	0.3583
1.1113	15.0	5625	1.2362	0.3617
1.1197	16.0	6000	1.2307	0.3633
1.1175	17.0	6375	1.2251	0.3667
1.1137	18.0	6750	1.2199	0.3683
1.1317	19.0	7125	1.2149	0.3667
1.0985	20.0	7500	1.2099	0.37
1.1037	21.0	7875	1.2054	0.37
1.1051	22.0	8250	1.2008	0.3717
1.1012	23.0	8625	1.1965	0.3717
1.0418	24.0	9000	1.1925	0.375
1.0922	25.0	9375	1.1886	0.3767
1.0809	26.0	9750	1.1848	0.3767
1.096	27.0	10125	1.1812	0.3767
1.0328	28.0	10500	1.1778	0.375
1.0501	29.0	10875	1.1745	0.3767
1.065	30.0	11250	1.1714	0.3767
1.0717	31.0	11625	1.1685	0.3783
1.104	32.0	12000	1.1658	0.3767
1.0567	33.0	12375	1.1632	0.3783
1.0632	34.0	12750	1.1607	0.3783
1.0635	35.0	13125	1.1585	0.3783
1.0477	36.0	13500	1.1563	0.3833
1.0721	37.0	13875	1.1544	0.385
1.0594	38.0	14250	1.1526	0.385
1.0484	39.0	14625	1.1510	0.3867
1.0408	40.0	15000	1.1495	0.3883
1.0421	41.0	15375	1.1482	0.39
1.0561	42.0	15750	1.1470	0.3883
1.0338	43.0	16125	1.1460	0.3883
1.0224	44.0	16500	1.1451	0.3883
1.0269	45.0	16875	1.1444	0.3883
1.0608	46.0	17250	1.1438	0.3883
1.0652	47.0	17625	1.1434	0.3883
1.0189	48.0	18000	1.1431	0.3883
1.0225	49.0	18375	1.1429	0.3883
1.0356	50.0	18750	1.1429	0.3883

Framework versions

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