beit-base-patch16-224-hasta-85-fold5

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: 1.6354
• Accuracy: 0.8182

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	1	1.0103	0.5455
No log	2.0	2	0.8046	0.6364
No log	3.0	3	0.7378	0.7273
No log	4.0	4	1.0676	0.7273
No log	5.0	5	1.3956	0.7273
No log	6.0	6	1.5799	0.7273
No log	7.0	7	1.6061	0.7273
No log	8.0	8	1.4351	0.7273
No log	9.0	9	1.3296	0.7273
0.39	10.0	10	1.2816	0.7273
0.39	11.0	11	1.2655	0.7273
0.39	12.0	12	1.2040	0.7273
0.39	13.0	13	1.0664	0.7273
0.39	14.0	14	1.0846	0.7273
0.39	15.0	15	1.2145	0.7273
0.39	16.0	16	1.4682	0.7273
0.39	17.0	17	1.4473	0.7273
0.39	18.0	18	1.2699	0.7273
0.39	19.0	19	1.2467	0.7273
0.1832	20.0	20	1.2648	0.7273
0.1832	21.0	21	1.2914	0.7273
0.1832	22.0	22	1.3444	0.7273
0.1832	23.0	23	1.5325	0.7273
0.1832	24.0	24	1.6140	0.7273
0.1832	25.0	25	1.6262	0.7273
0.1832	26.0	26	1.6753	0.7273
0.1832	27.0	27	1.6531	0.7273
0.1832	28.0	28	1.7096	0.7273
0.1832	29.0	29	1.6662	0.7273
0.1194	30.0	30	1.5769	0.7273
0.1194	31.0	31	1.4447	0.7273
0.1194	32.0	32	1.2644	0.7273
0.1194	33.0	33	1.2033	0.7273
0.1194	34.0	34	1.2703	0.7273
0.1194	35.0	35	1.4492	0.7273
0.1194	36.0	36	1.5890	0.7273
0.1194	37.0	37	1.5691	0.7273
0.1194	38.0	38	1.4127	0.7273
0.1194	39.0	39	1.3179	0.7273
0.0783	40.0	40	1.2986	0.7273
0.0783	41.0	41	1.3181	0.7273
0.0783	42.0	42	1.4253	0.7273
0.0783	43.0	43	1.5179	0.7273
0.0783	44.0	44	1.5685	0.7273
0.0783	45.0	45	1.5696	0.7273
0.0783	46.0	46	1.7571	0.7273
0.0783	47.0	47	1.9122	0.7273
0.0783	48.0	48	2.1062	0.7273
0.0783	49.0	49	2.1661	0.7273
0.056	50.0	50	2.1833	0.7273
0.056	51.0	51	2.2402	0.7273
0.056	52.0	52	2.3007	0.7273
0.056	53.0	53	2.3692	0.7273
0.056	54.0	54	2.3821	0.7273
0.056	55.0	55	2.2716	0.7273
0.056	56.0	56	2.0482	0.7273
0.056	57.0	57	1.8783	0.7273
0.056	58.0	58	1.7967	0.7273
0.056	59.0	59	1.7036	0.7273
0.052	60.0	60	1.6389	0.7273
0.052	61.0	61	1.6354	0.8182
0.052	62.0	62	1.6852	0.8182
0.052	63.0	63	1.8189	0.7273
0.052	64.0	64	1.9683	0.7273
0.052	65.0	65	2.0166	0.7273
0.052	66.0	66	2.0855	0.7273
0.052	67.0	67	2.1359	0.7273
0.052	68.0	68	2.2465	0.7273
0.052	69.0	69	2.2680	0.7273
0.0276	70.0	70	2.2728	0.7273
0.0276	71.0	71	2.2820	0.7273
0.0276	72.0	72	2.2427	0.7273
0.0276	73.0	73	2.2066	0.7273
0.0276	74.0	74	2.2434	0.7273
0.0276	75.0	75	2.3206	0.7273
0.0276	76.0	76	2.4408	0.7273
0.0276	77.0	77	2.4810	0.7273
0.0276	78.0	78	2.5091	0.7273
0.0276	79.0	79	2.4862	0.7273
0.0411	80.0	80	2.4502	0.7273
0.0411	81.0	81	2.4204	0.7273
0.0411	82.0	82	2.3838	0.7273
0.0411	83.0	83	2.3431	0.7273
0.0411	84.0	84	2.2927	0.7273
0.0411	85.0	85	2.2181	0.7273
0.0411	86.0	86	2.1633	0.7273
0.0411	87.0	87	2.0966	0.7273
0.0411	88.0	88	2.0536	0.7273
0.0411	89.0	89	2.0427	0.7273
0.0317	90.0	90	2.0524	0.7273
0.0317	91.0	91	2.0489	0.7273
0.0317	92.0	92	2.0648	0.7273
0.0317	93.0	93	2.0946	0.7273
0.0317	94.0	94	2.1155	0.7273
0.0317	95.0	95	2.1469	0.7273
0.0317	96.0	96	2.1768	0.7273
0.0317	97.0	97	2.2026	0.7273
0.0317	98.0	98	2.2205	0.7273
0.0317	99.0	99	2.2304	0.7273
0.0394	100.0	100	2.2350	0.7273

Framework versions

• Transformers 4.41.0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1