vit-base-patch16-224-Trial008-YEL_STEM2

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: 0.1024
• 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
0.7598	0.67	1	0.6968	0.6471
0.7294	2.0	3	0.7662	0.4706
0.6662	2.67	4	0.7196	0.5882
0.5662	4.0	6	0.3941	0.8235
0.4781	4.67	7	0.3458	0.8235
0.3259	6.0	9	0.1699	0.9412
0.2903	6.67	10	0.1024	1.0
0.2206	8.0	12	0.0788	1.0
0.3215	8.67	13	0.0414	1.0
0.1741	10.0	15	0.0218	1.0
0.2222	10.67	16	0.0207	1.0
0.1534	12.0	18	0.0128	1.0
0.273	12.67	19	0.0103	1.0
0.2021	14.0	21	0.0080	1.0
0.2193	14.67	22	0.0100	1.0
0.2132	16.0	24	0.0247	1.0
0.2163	16.67	25	0.0266	1.0
0.1626	18.0	27	0.0101	1.0
0.2492	18.67	28	0.0059	1.0
0.1308	20.0	30	0.0056	1.0
0.2144	20.67	31	0.0060	1.0
0.1389	22.0	33	0.0044	1.0
0.2548	22.67	34	0.0040	1.0
0.1324	24.0	36	0.0037	1.0
0.1958	24.67	37	0.0036	1.0
0.2476	26.0	39	0.0035	1.0
0.1439	26.67	40	0.0033	1.0
0.1202	28.0	42	0.0030	1.0
0.1368	28.67	43	0.0028	1.0
0.1016	30.0	45	0.0027	1.0
0.1282	30.67	46	0.0027	1.0
0.1128	32.0	48	0.0026	1.0
0.2366	32.67	49	0.0026	1.0
0.1727	33.33	50	0.0026	1.0

Framework versions

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