smids_5x_deit_base_rms_0001_fold4

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: 1.4272
• Accuracy: 0.885

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
0.2738	1.0	375	0.3688	0.8617
0.164	2.0	750	0.4071	0.86
0.1224	3.0	1125	0.5148	0.87
0.1153	4.0	1500	0.5755	0.8833
0.0768	5.0	1875	0.5201	0.8733
0.0128	6.0	2250	0.6962	0.88
0.0434	7.0	2625	0.7802	0.8733
0.011	8.0	3000	0.8928	0.8567
0.0054	9.0	3375	0.8190	0.8583
0.0033	10.0	3750	0.8366	0.8683
0.0196	11.0	4125	0.7608	0.8817
0.0229	12.0	4500	0.8098	0.8683
0.032	13.0	4875	0.8776	0.865
0.0132	14.0	5250	0.9792	0.8517
0.0019	15.0	5625	0.8844	0.8783
0.0007	16.0	6000	0.9761	0.8683
0.0282	17.0	6375	0.8085	0.855
0.0269	18.0	6750	0.8221	0.875
0.0	19.0	7125	0.8222	0.8833
0.0006	20.0	7500	0.9491	0.8633
0.0061	21.0	7875	0.9907	0.86
0.0014	22.0	8250	1.0614	0.86
0.0136	23.0	8625	0.8637	0.8717
0.0038	24.0	9000	0.9073	0.8717
0.003	25.0	9375	1.0178	0.875
0.0311	26.0	9750	0.9666	0.8817
0.0228	27.0	10125	0.9904	0.8783
0.0	28.0	10500	1.1423	0.8617
0.0001	29.0	10875	1.1415	0.865
0.0142	30.0	11250	1.0575	0.88
0.0005	31.0	11625	1.2704	0.8683
0.0	32.0	12000	1.1747	0.875
0.0055	33.0	12375	1.1305	0.8717
0.0001	34.0	12750	1.1488	0.8817
0.0	35.0	13125	1.1227	0.8767
0.0	36.0	13500	1.1854	0.8733
0.0	37.0	13875	1.1866	0.8817
0.0066	38.0	14250	1.1780	0.885
0.0	39.0	14625	1.2908	0.88
0.0	40.0	15000	1.2885	0.8817
0.0	41.0	15375	1.3508	0.885
0.0	42.0	15750	1.3636	0.8817
0.0	43.0	16125	1.3897	0.8817
0.0	44.0	16500	1.3912	0.8783
0.0	45.0	16875	1.4003	0.885
0.0	46.0	17250	1.4104	0.885
0.0	47.0	17625	1.4228	0.885
0.0	48.0	18000	1.4261	0.885
0.0	49.0	18375	1.4274	0.885
0.0	50.0	18750	1.4272	0.885

Framework versions

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