smids_10x_deit_base_sgd_00001_fold5

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: 1.0147
• Accuracy: 0.53

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
1.1118	1.0	750	1.1081	0.3483
1.0989	2.0	1500	1.1045	0.3467
1.098	3.0	2250	1.1011	0.3483
1.0956	4.0	3000	1.0977	0.35
1.0908	5.0	3750	1.0945	0.3533
1.0728	6.0	4500	1.0914	0.36
1.0863	7.0	5250	1.0883	0.3783
1.0893	8.0	6000	1.0853	0.3833
1.0669	9.0	6750	1.0824	0.3883
1.0847	10.0	7500	1.0795	0.3933
1.062	11.0	8250	1.0767	0.4033
1.0612	12.0	9000	1.0739	0.4183
1.0546	13.0	9750	1.0712	0.4267
1.0486	14.0	10500	1.0685	0.43
1.0488	15.0	11250	1.0659	0.4367
1.0504	16.0	12000	1.0633	0.4367
1.0424	17.0	12750	1.0607	0.4417
1.0426	18.0	13500	1.0582	0.4533
1.0273	19.0	14250	1.0557	0.4533
1.0365	20.0	15000	1.0533	0.4633
1.0042	21.0	15750	1.0509	0.4683
1.0296	22.0	16500	1.0486	0.4717
1.0417	23.0	17250	1.0462	0.475
1.0189	24.0	18000	1.0440	0.4817
1.0029	25.0	18750	1.0418	0.4833
1.0163	26.0	19500	1.0396	0.4883
1.0097	27.0	20250	1.0376	0.4867
1.0284	28.0	21000	1.0356	0.49
1.0024	29.0	21750	1.0337	0.4967
1.0018	30.0	22500	1.0318	0.505
1.0059	31.0	23250	1.0301	0.5083
1.0064	32.0	24000	1.0284	0.5133
1.0048	33.0	24750	1.0269	0.5133
0.9972	34.0	25500	1.0254	0.5217
0.9969	35.0	26250	1.0240	0.5217
1.0105	36.0	27000	1.0228	0.5217
0.9858	37.0	27750	1.0216	0.525
1.0046	38.0	28500	1.0205	0.525
0.9725	39.0	29250	1.0195	0.5267
0.9836	40.0	30000	1.0186	0.5267
0.9756	41.0	30750	1.0178	0.5283
0.9954	42.0	31500	1.0171	0.5283
1.0089	43.0	32250	1.0165	0.5283
0.9937	44.0	33000	1.0160	0.5267
0.9919	45.0	33750	1.0156	0.53
0.9943	46.0	34500	1.0153	0.53
0.9683	47.0	35250	1.0150	0.53
0.9833	48.0	36000	1.0148	0.53
0.9815	49.0	36750	1.0147	0.53
0.9749	50.0	37500	1.0147	0.53

Framework versions

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