smids_10x_deit_small_rms_0001_fold3

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

• Loss: 1.0751
• Accuracy: 0.905

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
0.2232	1.0	750	0.2644	0.9033
0.1749	2.0	1500	0.3888	0.87
0.0708	3.0	2250	0.3860	0.9083
0.0864	4.0	3000	0.4018	0.9017
0.027	5.0	3750	0.5540	0.905
0.0152	6.0	4500	0.5713	0.9133
0.0637	7.0	5250	0.5356	0.9017
0.0914	8.0	6000	0.5134	0.9117
0.0042	9.0	6750	0.5226	0.905
0.0385	10.0	7500	0.6176	0.9067
0.0198	11.0	8250	0.7633	0.89
0.0448	12.0	9000	0.7337	0.8783
0.0021	13.0	9750	0.6859	0.8917
0.0019	14.0	10500	0.7372	0.8933
0.0006	15.0	11250	0.7460	0.885
0.0195	16.0	12000	0.7805	0.8933
0.0202	17.0	12750	0.8243	0.895
0.016	18.0	13500	0.7845	0.89
0.0037	19.0	14250	0.7538	0.8883
0.0001	20.0	15000	0.6925	0.8967
0.0006	21.0	15750	0.8393	0.8933
0.0	22.0	16500	0.7236	0.9
0.0024	23.0	17250	0.8639	0.885
0.0014	24.0	18000	0.8799	0.8917
0.0236	25.0	18750	0.6893	0.9033
0.0001	26.0	19500	0.7435	0.9033
0.0001	27.0	20250	0.6829	0.89
0.0194	28.0	21000	0.8267	0.8967
0.0002	29.0	21750	0.8000	0.8983
0.0001	30.0	22500	0.8336	0.89
0.0	31.0	23250	0.8017	0.9
0.0	32.0	24000	0.8257	0.9117
0.0	33.0	24750	0.8456	0.905
0.0	34.0	25500	0.7637	0.91
0.0	35.0	26250	0.8426	0.9067
0.0219	36.0	27000	0.8594	0.9067
0.0	37.0	27750	0.8437	0.9083
0.0	38.0	28500	0.9026	0.9117
0.0	39.0	29250	0.9566	0.9067
0.0	40.0	30000	0.9200	0.915
0.0	41.0	30750	0.9067	0.92
0.0	42.0	31500	0.9289	0.91
0.0	43.0	32250	0.9815	0.91
0.0	44.0	33000	0.9712	0.91
0.0	45.0	33750	1.0254	0.9067
0.0	46.0	34500	1.0353	0.9083
0.0	47.0	35250	1.0450	0.905
0.0	48.0	36000	1.0661	0.905
0.0	49.0	36750	1.0715	0.905
0.0	50.0	37500	1.0751	0.905

Framework versions

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