smids_10x_deit_base_sgd_001_fold3

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

• Loss: 0.2607
• Accuracy: 0.9083

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: 0.001
• 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
0.5671	1.0	750	0.5928	0.7967
0.408	2.0	1500	0.4283	0.8517
0.3447	3.0	2250	0.3742	0.8633
0.3088	4.0	3000	0.3475	0.8683
0.2979	5.0	3750	0.3269	0.8733
0.2962	6.0	4500	0.3183	0.8767
0.2557	7.0	5250	0.3059	0.8817
0.2555	8.0	6000	0.2957	0.8817
0.2367	9.0	6750	0.2914	0.885
0.1949	10.0	7500	0.2859	0.8917
0.2488	11.0	8250	0.2846	0.8917
0.2475	12.0	9000	0.2777	0.895
0.1828	13.0	9750	0.2753	0.8983
0.2439	14.0	10500	0.2718	0.9017
0.2473	15.0	11250	0.2704	0.895
0.1928	16.0	12000	0.2696	0.895
0.1843	17.0	12750	0.2710	0.8983
0.2029	18.0	13500	0.2625	0.9033
0.2035	19.0	14250	0.2665	0.9
0.1744	20.0	15000	0.2677	0.905
0.152	21.0	15750	0.2612	0.9033
0.1898	22.0	16500	0.2631	0.9
0.1983	23.0	17250	0.2648	0.9067
0.1746	24.0	18000	0.2651	0.9067
0.2045	25.0	18750	0.2633	0.9067
0.1969	26.0	19500	0.2578	0.9067
0.1227	27.0	20250	0.2593	0.91
0.1518	28.0	21000	0.2610	0.9083
0.1661	29.0	21750	0.2607	0.9067
0.1698	30.0	22500	0.2600	0.9083
0.1513	31.0	23250	0.2624	0.9067
0.1181	32.0	24000	0.2595	0.9083
0.1772	33.0	24750	0.2601	0.9083
0.1745	34.0	25500	0.2608	0.9083
0.1241	35.0	26250	0.2599	0.9083
0.1459	36.0	27000	0.2607	0.9083
0.1333	37.0	27750	0.2607	0.9083
0.1934	38.0	28500	0.2605	0.905
0.1357	39.0	29250	0.2594	0.91
0.1781	40.0	30000	0.2597	0.91
0.1473	41.0	30750	0.2601	0.91
0.1773	42.0	31500	0.2596	0.9083
0.1488	43.0	32250	0.2600	0.9083
0.1451	44.0	33000	0.2615	0.9083
0.1365	45.0	33750	0.2606	0.9083
0.1144	46.0	34500	0.2619	0.9067
0.1714	47.0	35250	0.2607	0.9083
0.1189	48.0	36000	0.2607	0.9083
0.1679	49.0	36750	0.2607	0.9083
0.1606	50.0	37500	0.2607	0.9083

Framework versions

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