beit-base-patch16-224-hasta-85-fold2

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: 1.0030
• Accuracy: 0.7273

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	1.3943	0.0
No log	2.0	2	1.1771	0.3636
No log	3.0	3	1.0030	0.7273
No log	4.0	4	1.1175	0.7273
No log	5.0	5	1.3271	0.7273
No log	6.0	6	1.3905	0.7273
No log	7.0	7	1.2948	0.7273
No log	8.0	8	1.0699	0.7273
No log	9.0	9	0.9284	0.7273
0.3023	10.0	10	0.9573	0.7273
0.3023	11.0	11	1.1350	0.7273
0.3023	12.0	12	1.2566	0.7273
0.3023	13.0	13	1.2979	0.7273
0.3023	14.0	14	1.1942	0.7273
0.3023	15.0	15	1.1980	0.7273
0.3023	16.0	16	1.2017	0.7273
0.3023	17.0	17	1.4194	0.7273
0.3023	18.0	18	1.5204	0.7273
0.3023	19.0	19	1.3899	0.7273
0.1701	20.0	20	1.2407	0.7273
0.1701	21.0	21	1.3356	0.7273
0.1701	22.0	22	1.5076	0.7273
0.1701	23.0	23	1.4260	0.7273
0.1701	24.0	24	1.1877	0.7273
0.1701	25.0	25	1.0433	0.7273
0.1701	26.0	26	1.0261	0.7273
0.1701	27.0	27	1.0869	0.7273
0.1701	28.0	28	1.1074	0.7273
0.1701	29.0	29	1.0858	0.7273
0.1058	30.0	30	1.0020	0.7273
0.1058	31.0	31	0.9881	0.7273
0.1058	32.0	32	1.0530	0.7273
0.1058	33.0	33	1.3736	0.7273
0.1058	34.0	34	1.4768	0.7273
0.1058	35.0	35	1.4372	0.7273
0.1058	36.0	36	1.4594	0.7273
0.1058	37.0	37	1.4529	0.7273
0.1058	38.0	38	1.6027	0.7273
0.1058	39.0	39	1.7376	0.7273
0.065	40.0	40	1.8993	0.7273
0.065	41.0	41	1.9927	0.7273
0.065	42.0	42	1.8867	0.7273
0.065	43.0	43	1.6363	0.7273
0.065	44.0	44	1.5642	0.7273
0.065	45.0	45	1.5278	0.7273
0.065	46.0	46	1.5097	0.7273
0.065	47.0	47	1.5586	0.7273
0.065	48.0	48	1.5659	0.7273
0.065	49.0	49	1.5743	0.7273
0.061	50.0	50	1.5951	0.7273
0.061	51.0	51	1.6097	0.7273
0.061	52.0	52	1.6781	0.7273
0.061	53.0	53	1.7168	0.7273
0.061	54.0	54	1.6331	0.7273
0.061	55.0	55	1.5711	0.7273
0.061	56.0	56	1.6043	0.7273
0.061	57.0	57	1.6590	0.7273
0.061	58.0	58	1.6879	0.7273
0.061	59.0	59	1.6452	0.7273
0.0642	60.0	60	1.6099	0.7273
0.0642	61.0	61	1.5536	0.7273
0.0642	62.0	62	1.5496	0.7273
0.0642	63.0	63	1.5528	0.7273
0.0642	64.0	64	1.6351	0.7273
0.0642	65.0	65	1.7556	0.7273
0.0642	66.0	66	1.8993	0.7273
0.0642	67.0	67	2.0309	0.7273
0.0642	68.0	68	2.1548	0.7273
0.0642	69.0	69	2.2087	0.7273
0.0411	70.0	70	2.2062	0.7273
0.0411	71.0	71	2.1605	0.7273
0.0411	72.0	72	2.1347	0.7273
0.0411	73.0	73	2.0662	0.7273
0.0411	74.0	74	2.0683	0.7273
0.0411	75.0	75	2.0466	0.7273
0.0411	76.0	76	1.9756	0.7273
0.0411	77.0	77	1.8928	0.7273
0.0411	78.0	78	1.8972	0.7273
0.0411	79.0	79	1.9408	0.7273
0.0421	80.0	80	1.9690	0.7273
0.0421	81.0	81	2.0466	0.7273
0.0421	82.0	82	2.1174	0.7273
0.0421	83.0	83	2.1825	0.7273
0.0421	84.0	84	2.2527	0.7273
0.0421	85.0	85	2.2933	0.7273
0.0421	86.0	86	2.3311	0.7273
0.0421	87.0	87	2.3468	0.7273
0.0421	88.0	88	2.3222	0.7273
0.0421	89.0	89	2.2764	0.7273
0.0304	90.0	90	2.2190	0.7273
0.0304	91.0	91	2.1855	0.7273
0.0304	92.0	92	2.1677	0.7273
0.0304	93.0	93	2.1493	0.7273
0.0304	94.0	94	2.1259	0.7273
0.0304	95.0	95	2.1151	0.7273
0.0304	96.0	96	2.1179	0.7273
0.0304	97.0	97	2.1250	0.7273
0.0304	98.0	98	2.1302	0.7273
0.0304	99.0	99	2.1330	0.7273
0.0305	100.0	100	2.1360	0.7273

Framework versions

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