smids_5x_deit_small_rms_00001_fold2

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: 1.3142
• Accuracy: 0.8636

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
0.255	1.0	375	0.2860	0.8785
0.192	2.0	750	0.3380	0.8719
0.111	3.0	1125	0.5269	0.8536
0.0357	4.0	1500	0.6082	0.8819
0.0344	5.0	1875	0.7662	0.8586
0.0306	6.0	2250	0.7220	0.8735
0.0004	7.0	2625	1.0133	0.8636
0.0601	8.0	3000	0.9769	0.8602
0.0002	9.0	3375	1.0509	0.8719
0.0001	10.0	3750	1.0508	0.8686
0.0242	11.0	4125	1.1405	0.8619
0.0086	12.0	4500	0.9578	0.8735
0.0	13.0	4875	0.9452	0.8702
0.0167	14.0	5250	1.1793	0.8652
0.0068	15.0	5625	1.1314	0.8569
0.0242	16.0	6000	1.0830	0.8686
0.0116	17.0	6375	1.0898	0.8686
0.0019	18.0	6750	1.1516	0.8702
0.0	19.0	7125	1.1246	0.8686
0.0357	20.0	7500	1.2754	0.8419
0.0167	21.0	7875	1.1083	0.8586
0.0	22.0	8250	1.1597	0.8636
0.0	23.0	8625	1.1775	0.8686
0.0001	24.0	9000	1.1781	0.8735
0.0	25.0	9375	1.1367	0.8752
0.0	26.0	9750	1.2570	0.8602
0.0	27.0	10125	1.2344	0.8669
0.0	28.0	10500	1.2212	0.8686
0.0	29.0	10875	1.1884	0.8686
0.0	30.0	11250	1.1717	0.8819
0.0041	31.0	11625	1.2327	0.8735
0.0049	32.0	12000	1.2073	0.8719
0.0069	33.0	12375	1.2981	0.8669
0.0	34.0	12750	1.3346	0.8602
0.0	35.0	13125	1.2237	0.8719
0.0	36.0	13500	1.2742	0.8702
0.0	37.0	13875	1.3127	0.8702
0.0	38.0	14250	1.3037	0.8702
0.0	39.0	14625	1.3578	0.8636
0.0033	40.0	15000	1.3159	0.8636
0.0	41.0	15375	1.3110	0.8686
0.0024	42.0	15750	1.3216	0.8652
0.0026	43.0	16125	1.3041	0.8686
0.0026	44.0	16500	1.3057	0.8669
0.0024	45.0	16875	1.3146	0.8652
0.0	46.0	17250	1.3144	0.8686
0.0053	47.0	17625	1.3156	0.8652
0.0	48.0	18000	1.3154	0.8652
0.0023	49.0	18375	1.3139	0.8652
0.0024	50.0	18750	1.3142	0.8636

Framework versions

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