vit-base-patch16-224-Trial006_007_008-YEL_STEM1

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

• Loss: 0.0453
• Accuracy: 1.0

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: 5e-05
• train_batch_size: 30
• eval_batch_size: 30
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 120
• 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
0.7064	1.0	5	0.6605	0.5373
0.5838	2.0	10	0.5289	0.8358
0.4909	3.0	15	0.3967	0.8209
0.3317	4.0	20	0.2759	0.9104
0.2813	5.0	25	0.1820	0.9403
0.2948	6.0	30	0.1286	0.9552
0.2253	7.0	35	0.0453	1.0
0.2125	8.0	40	0.0408	1.0
0.1288	9.0	45	0.0177	1.0
0.1084	10.0	50	0.0265	1.0
0.1642	11.0	55	0.0994	0.9403
0.149	12.0	60	0.0316	0.9851
0.1315	13.0	65	0.0325	0.9851
0.1101	14.0	70	0.1090	0.9701
0.1101	15.0	75	0.0094	1.0
0.0702	16.0	80	0.0070	1.0
0.1184	17.0	85	0.0634	0.9851
0.1506	18.0	90	0.0104	1.0
0.1027	19.0	95	0.0149	1.0
0.159	20.0	100	0.1021	0.9552
0.1205	21.0	105	0.0085	1.0
0.1511	22.0	110	0.0248	0.9851
0.2228	23.0	115	0.0993	0.9552
0.1431	24.0	120	0.0373	0.9851
0.1489	25.0	125	0.0161	1.0
0.0799	26.0	130	0.0382	0.9851
0.1411	27.0	135	0.0071	1.0
0.1457	28.0	140	0.0047	1.0
0.1434	29.0	145	0.0069	1.0
0.0913	30.0	150	0.0032	1.0
0.1354	31.0	155	0.0042	1.0
0.1253	32.0	160	0.0061	1.0
0.1065	33.0	165	0.0039	1.0
0.1199	34.0	170	0.0023	1.0
0.1274	35.0	175	0.0037	1.0
0.1118	36.0	180	0.0100	1.0
0.1237	37.0	185	0.0053	1.0
0.1311	38.0	190	0.0028	1.0
0.1833	39.0	195	0.0026	1.0
0.0858	40.0	200	0.0033	1.0
0.1503	41.0	205	0.0049	1.0
0.0547	42.0	210	0.0037	1.0
0.1647	43.0	215	0.0037	1.0
0.1066	44.0	220	0.0061	1.0
0.1277	45.0	225	0.0083	1.0
0.0885	46.0	230	0.0083	1.0
0.1339	47.0	235	0.0081	1.0
0.0904	48.0	240	0.0073	1.0
0.079	49.0	245	0.0080	1.0
0.0788	50.0	250	0.0086	1.0

Framework versions

• Transformers 4.30.0.dev0
• Pytorch 1.12.1
• Datasets 2.12.0
• Tokenizers 0.13.1