vit-base-patch16-224-Trial008-YEL_STEM4

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: 0.2608
• 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	0.8	1	0.7803	0.4667
0.6891	1.6	2	0.6427	0.6
0.6152	2.4	3	0.5351	0.6667
0.6156	4.0	5	0.3241	0.8667
0.4845	4.8	6	0.2608	1.0
0.4413	5.6	7	0.1876	1.0
0.4024	6.4	8	0.1277	1.0
0.3111	8.0	10	0.1968	0.9333
0.3655	8.8	11	0.0837	1.0
0.3985	9.6	12	0.0476	1.0
0.2309	10.4	13	0.0446	1.0
0.2046	12.0	15	0.0307	1.0
0.2295	12.8	16	0.0220	1.0
0.1953	13.6	17	0.0161	1.0
0.1944	14.4	18	0.0135	1.0
0.256	16.0	20	0.0097	1.0
0.2346	16.8	21	0.0081	1.0
0.1702	17.6	22	0.0073	1.0
0.1831	18.4	23	0.0067	1.0
0.2229	20.0	25	0.0061	1.0
0.1109	20.8	26	0.0053	1.0
0.292	21.6	27	0.0046	1.0
0.1701	22.4	28	0.0040	1.0
0.1471	24.0	30	0.0031	1.0
0.1796	24.8	31	0.0028	1.0
0.1232	25.6	32	0.0026	1.0
0.1646	26.4	33	0.0024	1.0
0.1649	28.0	35	0.0021	1.0
0.1294	28.8	36	0.0021	1.0
0.1352	29.6	37	0.0020	1.0
0.0985	30.4	38	0.0020	1.0
0.1209	32.0	40	0.0020	1.0
0.1004	32.8	41	0.0020	1.0
0.0718	33.6	42	0.0019	1.0
0.1028	34.4	43	0.0019	1.0
0.0984	36.0	45	0.0018	1.0
0.0866	36.8	46	0.0018	1.0
0.136	37.6	47	0.0018	1.0
0.1433	38.4	48	0.0018	1.0
0.1121	40.0	50	0.0017	1.0

Framework versions

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