vit-base-patch16-224-Trial006-YEL_STEM

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: 0.0568
• 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
0.6941	1.0	2	0.7689	0.5385
0.6362	2.0	4	0.7104	0.5385
0.5959	3.0	6	0.6233	0.5769
0.5411	4.0	8	0.5129	0.6923
0.4225	5.0	10	0.4618	0.7692
0.3148	6.0	12	0.3855	0.8077
0.323	7.0	14	0.3392	0.8462
0.2525	8.0	16	0.3819	0.8462
0.3571	9.0	18	0.3386	0.8846
0.2474	10.0	20	0.3052	0.8462
0.2764	11.0	22	0.2868	0.9231
0.268	12.0	24	0.2079	0.9231
0.1943	13.0	26	0.1218	0.9615
0.225	14.0	28	0.0912	0.9615
0.2672	15.0	30	0.0771	0.9615
0.1485	16.0	32	0.0568	1.0
0.278	17.0	34	0.0365	1.0
0.1887	18.0	36	0.1186	0.9231
0.2053	19.0	38	0.1227	0.9231
0.1519	20.0	40	0.0994	0.9615
0.1435	21.0	42	0.2631	0.8846
0.2232	22.0	44	0.2108	0.9231
0.1737	23.0	46	0.0582	1.0
0.2007	24.0	48	0.0550	1.0
0.1747	25.0	50	0.0307	1.0
0.1821	26.0	52	0.0976	0.9231
0.2866	27.0	54	0.0281	1.0
0.1574	28.0	56	0.0176	1.0
0.1835	29.0	58	0.0731	0.9615
0.1768	30.0	60	0.1153	0.9615
0.1916	31.0	62	0.0964	0.9615
0.1383	32.0	64	0.0766	0.9615
0.0834	33.0	66	0.0758	0.9231
0.2194	34.0	68	0.0392	1.0
0.1497	35.0	70	0.0182	1.0
0.1891	36.0	72	0.0167	1.0
0.2006	37.0	74	0.0097	1.0
0.1414	38.0	76	0.0077	1.0
0.2464	39.0	78	0.0101	1.0
0.1928	40.0	80	0.0074	1.0
0.1269	41.0	82	0.0054	1.0
0.1622	42.0	84	0.0049	1.0
0.1637	43.0	86	0.0057	1.0
0.2383	44.0	88	0.0054	1.0
0.1373	45.0	90	0.0049	1.0
0.1903	46.0	92	0.0048	1.0
0.1371	47.0	94	0.0048	1.0
0.2254	48.0	96	0.0048	1.0
0.1254	49.0	98	0.0048	1.0
0.1982	50.0	100	0.0049	1.0

Framework versions

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