smids_10x_deit_tiny_adamax_00001_fold4

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

• Loss: 1.3154
• Accuracy: 0.87

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
0.2882	1.0	750	0.3671	0.8467
0.2123	2.0	1500	0.3694	0.8633
0.189	3.0	2250	0.3920	0.8617
0.1612	4.0	3000	0.4004	0.8733
0.1046	5.0	3750	0.4710	0.85
0.0761	6.0	4500	0.4693	0.86
0.0433	7.0	5250	0.5028	0.8633
0.0491	8.0	6000	0.5468	0.865
0.0675	9.0	6750	0.5733	0.8717
0.039	10.0	7500	0.6491	0.875
0.0538	11.0	8250	0.7887	0.8567
0.012	12.0	9000	0.7915	0.87
0.0116	13.0	9750	0.8692	0.87
0.0189	14.0	10500	0.8744	0.8717
0.0002	15.0	11250	0.9425	0.8683
0.024	16.0	12000	0.9688	0.8633
0.0001	17.0	12750	0.9969	0.87
0.0001	18.0	13500	1.0452	0.865
0.0001	19.0	14250	1.0740	0.87
0.0	20.0	15000	1.0493	0.8733
0.0	21.0	15750	1.0980	0.87
0.0	22.0	16500	1.1024	0.8683
0.0	23.0	17250	1.1005	0.8733
0.0	24.0	18000	1.1084	0.875
0.0	25.0	18750	1.1253	0.865
0.0	26.0	19500	1.1772	0.87
0.0	27.0	20250	1.1989	0.87
0.0	28.0	21000	1.2090	0.87
0.0	29.0	21750	1.2004	0.8683
0.0	30.0	22500	1.2111	0.87
0.0	31.0	23250	1.2258	0.87
0.0	32.0	24000	1.2322	0.87
0.0	33.0	24750	1.2293	0.87
0.0	34.0	25500	1.2340	0.8683
0.0	35.0	26250	1.2422	0.8667
0.0	36.0	27000	1.2597	0.8667
0.0	37.0	27750	1.2593	0.8683
0.0	38.0	28500	1.2651	0.87
0.0	39.0	29250	1.2897	0.8667
0.0	40.0	30000	1.2906	0.87
0.0	41.0	30750	1.2961	0.87
0.0	42.0	31500	1.2976	0.8683
0.0	43.0	32250	1.2961	0.87
0.0	44.0	33000	1.3012	0.8683
0.0	45.0	33750	1.3074	0.87
0.0	46.0	34500	1.3114	0.8683
0.0	47.0	35250	1.3127	0.87
0.0	48.0	36000	1.3140	0.87
0.0	49.0	36750	1.3150	0.87
0.0	50.0	37500	1.3154	0.87

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.1+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2