vit-base-patch16-224-Trial006-YEL_STEM3

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: 0.0370
• 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: 60
• eval_batch_size: 60
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 240
• 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
0.7214	1.0	2	0.6865	0.625
0.6538	2.0	4	0.6506	0.6875
0.5973	3.0	6	0.5406	0.7708
0.5867	4.0	8	0.4452	0.8125
0.4882	5.0	10	0.3944	0.9167
0.3508	6.0	12	0.3255	0.8125
0.3367	7.0	14	0.2000	0.9375
0.2721	8.0	16	0.1377	0.9792
0.2401	9.0	18	0.0991	0.9792
0.242	10.0	20	0.0952	0.9583
0.2074	11.0	22	0.1278	0.9375
0.2048	12.0	24	0.0370	1.0
0.1933	13.0	26	0.1006	0.9375
0.1869	14.0	28	0.0348	1.0
0.2057	15.0	30	0.1574	0.9375
0.2368	16.0	32	0.0518	0.9792
0.1114	17.0	34	0.0149	1.0
0.1486	18.0	36	0.0187	1.0
0.1161	19.0	38	0.0083	1.0
0.1133	20.0	40	0.0062	1.0
0.1085	21.0	42	0.0108	1.0
0.1349	22.0	44	0.0148	1.0
0.1076	23.0	46	0.0080	1.0
0.1178	24.0	48	0.0137	1.0
0.1566	25.0	50	0.0074	1.0
0.1578	26.0	52	0.0064	1.0
0.1039	27.0	54	0.0077	1.0
0.1585	28.0	56	0.0058	1.0
0.1299	29.0	58	0.0130	1.0
0.1059	30.0	60	0.0075	1.0
0.1162	31.0	62	0.0151	1.0
0.1147	32.0	64	0.0100	1.0
0.1226	33.0	66	0.0581	0.9792
0.1264	34.0	68	0.1029	0.9792
0.0858	35.0	70	0.0594	0.9792
0.0671	36.0	72	0.0119	1.0
0.1381	37.0	74	0.0084	1.0
0.1054	38.0	76	0.0121	1.0
0.0969	39.0	78	0.0273	0.9792
0.1168	40.0	80	0.0203	0.9792
0.1065	41.0	82	0.0061	1.0
0.14	42.0	84	0.0041	1.0
0.1186	43.0	86	0.0088	1.0
0.0818	44.0	88	0.0214	0.9792
0.0676	45.0	90	0.0148	1.0
0.043	46.0	92	0.0105	1.0
0.0731	47.0	94	0.0085	1.0
0.1297	48.0	96	0.0082	1.0
0.1191	49.0	98	0.0086	1.0
0.0657	50.0	100	0.0090	1.0

Framework versions

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