beit-base-patch16-224-85-fold1

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

• Loss: 0.1430
• Accuracy: 0.9773

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	2	0.7308	0.5455
No log	2.0	4	0.7927	0.7045
No log	3.0	6	0.9672	0.7045
No log	4.0	8	0.6257	0.7045
0.6404	5.0	10	0.4646	0.7955
0.6404	6.0	12	0.5648	0.7045
0.6404	7.0	14	0.4389	0.7727
0.6404	8.0	16	0.4523	0.75
0.6404	9.0	18	0.4698	0.75
0.455	10.0	20	0.3707	0.8409
0.455	11.0	22	0.3594	0.8182
0.455	12.0	24	0.6136	0.7273
0.455	13.0	26	0.3022	0.8864
0.455	14.0	28	0.2919	0.8409
0.3981	15.0	30	0.3612	0.8182
0.3981	16.0	32	0.2492	0.8864
0.3981	17.0	34	0.2460	0.9091
0.3981	18.0	36	0.2931	0.8636
0.3981	19.0	38	0.1822	0.9091
0.3257	20.0	40	0.2060	0.9091
0.3257	21.0	42	0.2195	0.8864
0.3257	22.0	44	0.2624	0.9091
0.3257	23.0	46	0.2384	0.9091
0.3257	24.0	48	0.1767	0.9318
0.2553	25.0	50	0.2040	0.9318
0.2553	26.0	52	0.1981	0.9091
0.2553	27.0	54	0.1835	0.9318
0.2553	28.0	56	0.1820	0.9318
0.2553	29.0	58	0.1466	0.9545
0.2083	30.0	60	0.1668	0.9318
0.2083	31.0	62	0.2229	0.9318
0.2083	32.0	64	0.1783	0.9545
0.2083	33.0	66	0.1944	0.8864
0.2083	34.0	68	0.3025	0.9091
0.2353	35.0	70	0.4457	0.8409
0.2353	36.0	72	0.2759	0.9318
0.2353	37.0	74	0.2179	0.9318
0.2353	38.0	76	0.3911	0.9091
0.2353	39.0	78	0.5785	0.8409
0.1782	40.0	80	0.2339	0.9318
0.1782	41.0	82	0.2302	0.9091
0.1782	42.0	84	0.3967	0.8864
0.1782	43.0	86	0.4447	0.8636
0.1782	44.0	88	0.2020	0.9091
0.2059	45.0	90	0.1911	0.9318
0.2059	46.0	92	0.2609	0.9091
0.2059	47.0	94	0.2925	0.9091
0.2059	48.0	96	0.2079	0.9318
0.2059	49.0	98	0.1853	0.9318
0.1706	50.0	100	0.2860	0.9318
0.1706	51.0	102	0.3735	0.8636
0.1706	52.0	104	0.1968	0.9318
0.1706	53.0	106	0.1722	0.9318
0.1706	54.0	108	0.3123	0.8636
0.1429	55.0	110	0.3297	0.8864
0.1429	56.0	112	0.1430	0.9773
0.1429	57.0	114	0.1134	0.9773
0.1429	58.0	116	0.2312	0.9091
0.1429	59.0	118	0.2826	0.9091
0.1325	60.0	120	0.2417	0.9091
0.1325	61.0	122	0.1393	0.9318
0.1325	62.0	124	0.2178	0.9318
0.1325	63.0	126	0.3991	0.9091
0.1325	64.0	128	0.3325	0.9091
0.1481	65.0	130	0.2327	0.9091
0.1481	66.0	132	0.2885	0.9091
0.1481	67.0	134	0.3576	0.9091
0.1481	68.0	136	0.2686	0.9318
0.1481	69.0	138	0.1717	0.9545
0.1237	70.0	140	0.1493	0.9545
0.1237	71.0	142	0.1429	0.9318
0.1237	72.0	144	0.1790	0.9318
0.1237	73.0	146	0.1590	0.9318
0.1237	74.0	148	0.1971	0.8864
0.105	75.0	150	0.2229	0.9318
0.105	76.0	152	0.1789	0.8864
0.105	77.0	154	0.1671	0.9545
0.105	78.0	156	0.2435	0.9318
0.105	79.0	158	0.2658	0.9318
0.0923	80.0	160	0.2092	0.9318
0.0923	81.0	162	0.1748	0.9318
0.0923	82.0	164	0.1727	0.9318
0.0923	83.0	166	0.1945	0.9091
0.0923	84.0	168	0.2429	0.9318
0.1033	85.0	170	0.2796	0.9318
0.1033	86.0	172	0.2548	0.9318
0.1033	87.0	174	0.2379	0.9091
0.1033	88.0	176	0.2409	0.9091
0.1033	89.0	178	0.2421	0.9091
0.1073	90.0	180	0.2332	0.9091
0.1073	91.0	182	0.2231	0.9091
0.1073	92.0	184	0.2153	0.9318
0.1073	93.0	186	0.2088	0.9318
0.1073	94.0	188	0.2058	0.9318
0.104	95.0	190	0.2040	0.9318
0.104	96.0	192	0.2046	0.9318
0.104	97.0	194	0.2043	0.9318
0.104	98.0	196	0.2056	0.9318
0.104	99.0	198	0.2081	0.9318
0.0896	100.0	200	0.2097	0.9318

Framework versions

• Transformers 4.40.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1