smids_10x_deit_base_sgd_00001_fold3

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: 1.0056
• Accuracy: 0.5583

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
1.1044	1.0	750	1.1039	0.36
1.0936	2.0	1500	1.1001	0.3683
1.0833	3.0	2250	1.0965	0.385
1.0922	4.0	3000	1.0929	0.3867
1.0937	5.0	3750	1.0895	0.3933
1.0793	6.0	4500	1.0862	0.4083
1.0808	7.0	5250	1.0829	0.4133
1.0795	8.0	6000	1.0798	0.42
1.0735	9.0	6750	1.0767	0.4283
1.0874	10.0	7500	1.0736	0.435
1.0661	11.0	8250	1.0706	0.4483
1.038	12.0	9000	1.0677	0.445
1.0511	13.0	9750	1.0648	0.45
1.0539	14.0	10500	1.0620	0.4617
1.0408	15.0	11250	1.0592	0.46
1.0259	16.0	12000	1.0564	0.4617
1.0339	17.0	12750	1.0537	0.47
1.0416	18.0	13500	1.0511	0.4683
1.0423	19.0	14250	1.0484	0.4733
1.0333	20.0	15000	1.0459	0.4817
1.0189	21.0	15750	1.0434	0.485
1.0248	22.0	16500	1.0409	0.4917
1.0226	23.0	17250	1.0385	0.495
1.0228	24.0	18000	1.0362	0.5
1.0292	25.0	18750	1.0339	0.4983
1.02	26.0	19500	1.0317	0.5
1.0227	27.0	20250	1.0296	0.5017
1.0121	28.0	21000	1.0275	0.5083
1.0052	29.0	21750	1.0255	0.5183
0.9924	30.0	22500	1.0236	0.52
1.0074	31.0	23250	1.0218	0.5267
0.9858	32.0	24000	1.0200	0.5267
1.0028	33.0	24750	1.0184	0.5283
0.9845	34.0	25500	1.0169	0.5317
0.992	35.0	26250	1.0154	0.535
0.9814	36.0	27000	1.0141	0.535
1.0022	37.0	27750	1.0128	0.535
0.9973	38.0	28500	1.0117	0.5367
0.9952	39.0	29250	1.0106	0.54
0.9809	40.0	30000	1.0097	0.5417
0.9816	41.0	30750	1.0089	0.5467
1.0061	42.0	31500	1.0081	0.5533
0.9998	43.0	32250	1.0075	0.555
0.992	44.0	33000	1.0070	0.555
0.9773	45.0	33750	1.0065	0.5567
0.963	46.0	34500	1.0062	0.5567
0.9849	47.0	35250	1.0059	0.5567
0.9816	48.0	36000	1.0057	0.5583
0.999	49.0	36750	1.0056	0.5583
0.9748	50.0	37500	1.0056	0.5583

Framework versions

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