beit-base-patch16-224-hasta-85-fold4

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: 0.9258
• 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	1.0761	0.6364
No log	2.0	2	0.9258	0.7273
No log	3.0	3	0.8310	0.7273
No log	4.0	4	0.9402	0.7273
No log	5.0	5	1.1381	0.7273
No log	6.0	6	1.2812	0.7273
No log	7.0	7	1.2679	0.7273
No log	8.0	8	1.1704	0.7273
No log	9.0	9	1.1909	0.7273
0.3269	10.0	10	1.2981	0.7273
0.3269	11.0	11	1.2565	0.7273
0.3269	12.0	12	1.1475	0.7273
0.3269	13.0	13	1.0585	0.7273
0.3269	14.0	14	1.0294	0.7273
0.3269	15.0	15	1.0649	0.7273
0.3269	16.0	16	1.1712	0.7273
0.3269	17.0	17	1.2090	0.7273
0.3269	18.0	18	1.1579	0.7273
0.3269	19.0	19	1.0943	0.7273
0.1921	20.0	20	1.1877	0.7273
0.1921	21.0	21	1.3909	0.7273
0.1921	22.0	22	1.4301	0.7273
0.1921	23.0	23	1.4210	0.7273
0.1921	24.0	24	1.3994	0.7273
0.1921	25.0	25	1.3649	0.7273
0.1921	26.0	26	1.3244	0.7273
0.1921	27.0	27	1.2861	0.7273
0.1921	28.0	28	1.1634	0.7273
0.1921	29.0	29	0.9854	0.7273
0.1374	30.0	30	1.0608	0.7273
0.1374	31.0	31	1.3092	0.7273
0.1374	32.0	32	1.4679	0.7273
0.1374	33.0	33	1.4397	0.7273
0.1374	34.0	34	1.2949	0.7273
0.1374	35.0	35	1.2340	0.7273
0.1374	36.0	36	1.2524	0.7273
0.1374	37.0	37	1.2108	0.7273
0.1374	38.0	38	1.1878	0.7273
0.1374	39.0	39	1.1400	0.7273
0.0886	40.0	40	1.1186	0.7273
0.0886	41.0	41	1.3145	0.7273
0.0886	42.0	42	1.4749	0.7273
0.0886	43.0	43	1.5773	0.7273
0.0886	44.0	44	1.6792	0.7273
0.0886	45.0	45	1.7716	0.7273
0.0886	46.0	46	1.8943	0.7273
0.0886	47.0	47	1.8541	0.7273
0.0886	48.0	48	1.6656	0.7273
0.0886	49.0	49	1.4897	0.7273
0.0509	50.0	50	1.2921	0.7273
0.0509	51.0	51	1.2021	0.7273
0.0509	52.0	52	1.2643	0.7273
0.0509	53.0	53	1.4622	0.7273
0.0509	54.0	54	1.5043	0.7273
0.0509	55.0	55	1.5063	0.7273
0.0509	56.0	56	1.4604	0.7273
0.0509	57.0	57	1.3414	0.7273
0.0509	58.0	58	1.1789	0.7273
0.0509	59.0	59	1.1715	0.7273
0.0471	60.0	60	1.2550	0.7273
0.0471	61.0	61	1.3513	0.7273
0.0471	62.0	62	1.4922	0.7273
0.0471	63.0	63	1.6911	0.7273
0.0471	64.0	64	1.7747	0.7273
0.0471	65.0	65	1.7659	0.7273
0.0471	66.0	66	1.6730	0.7273
0.0471	67.0	67	1.5296	0.7273
0.0471	68.0	68	1.4973	0.7273
0.0471	69.0	69	1.4650	0.7273
0.0212	70.0	70	1.4970	0.7273
0.0212	71.0	71	1.5022	0.7273
0.0212	72.0	72	1.5275	0.7273
0.0212	73.0	73	1.5780	0.7273
0.0212	74.0	74	1.7149	0.7273
0.0212	75.0	75	1.8056	0.7273
0.0212	76.0	76	1.8394	0.7273
0.0212	77.0	77	1.8526	0.7273
0.0212	78.0	78	1.7944	0.7273
0.0212	79.0	79	1.7440	0.7273
0.0313	80.0	80	1.6994	0.7273
0.0313	81.0	81	1.6076	0.7273
0.0313	82.0	82	1.5753	0.7273
0.0313	83.0	83	1.5831	0.7273
0.0313	84.0	84	1.5471	0.7273
0.0313	85.0	85	1.5600	0.7273
0.0313	86.0	86	1.5832	0.7273
0.0313	87.0	87	1.5819	0.7273
0.0313	88.0	88	1.6053	0.7273
0.0313	89.0	89	1.6329	0.7273
0.0205	90.0	90	1.6751	0.7273
0.0205	91.0	91	1.6957	0.7273
0.0205	92.0	92	1.7326	0.7273
0.0205	93.0	93	1.7475	0.7273
0.0205	94.0	94	1.7503	0.7273
0.0205	95.0	95	1.7443	0.7273
0.0205	96.0	96	1.7483	0.7273
0.0205	97.0	97	1.7523	0.7273
0.0205	98.0	98	1.7516	0.7273
0.0205	99.0	99	1.7483	0.7273
0.0334	100.0	100	1.7462	0.7273

Framework versions

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