smids_10x_deit_tiny_rms_00001_fold3

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: 1.1108
• Accuracy: 0.8983

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
0.2467	1.0	750	0.2589	0.9033
0.1838	2.0	1500	0.3381	0.8833
0.1037	3.0	2250	0.3939	0.8817
0.0799	4.0	3000	0.4142	0.895
0.053	5.0	3750	0.4932	0.9
0.0314	6.0	4500	0.5727	0.8883
0.0421	7.0	5250	0.7156	0.8967
0.002	8.0	6000	0.7860	0.8983
0.001	9.0	6750	0.8016	0.9017
0.0004	10.0	7500	0.8581	0.8983
0.0041	11.0	8250	0.9068	0.8783
0.0193	12.0	9000	1.1367	0.875
0.0008	13.0	9750	0.9798	0.8917
0.0002	14.0	10500	1.0132	0.885
0.0001	15.0	11250	0.9596	0.9017
0.0	16.0	12000	0.9560	0.8967
0.0155	17.0	12750	0.9136	0.8967
0.0	18.0	13500	0.9761	0.8967
0.0	19.0	14250	1.1111	0.8983
0.0001	20.0	15000	1.0728	0.8933
0.0212	21.0	15750	0.9842	0.8917
0.0	22.0	16500	1.1318	0.88
0.0001	23.0	17250	0.9825	0.9
0.0	24.0	18000	1.1665	0.8917
0.0	25.0	18750	1.0643	0.8933
0.0	26.0	19500	1.0722	0.885
0.0	27.0	20250	1.1414	0.8867
0.0	28.0	21000	1.0530	0.89
0.0	29.0	21750	1.0157	0.89
0.0	30.0	22500	1.0643	0.8833
0.0001	31.0	23250	1.1945	0.88
0.0	32.0	24000	1.0688	0.885
0.0	33.0	24750	1.1471	0.8933
0.0001	34.0	25500	1.0005	0.8883
0.0	35.0	26250	1.0647	0.89
0.0	36.0	27000	1.0647	0.9
0.0	37.0	27750	1.1102	0.9
0.0	38.0	28500	1.0870	0.8967
0.0	39.0	29250	1.1068	0.89
0.0	40.0	30000	1.0975	0.9
0.0	41.0	30750	1.1055	0.895
0.0	42.0	31500	1.1130	0.9
0.0	43.0	32250	1.1059	0.8983
0.0	44.0	33000	1.1016	0.8967
0.0	45.0	33750	1.1080	0.8967
0.0	46.0	34500	1.1091	0.8967
0.0	47.0	35250	1.1121	0.8967
0.0	48.0	36000	1.1138	0.8983
0.0	49.0	36750	1.1129	0.8967
0.0	50.0	37500	1.1108	0.8983

Framework versions

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