beit-base-patch16-224-hasta-85-fold3

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: 0.8911
• 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	1.3301	0.1818
No log	2.0	2	1.1012	0.3636
No log	3.0	3	0.8911	0.7273
No log	4.0	4	0.9555	0.7273
No log	5.0	5	1.2582	0.7273
No log	6.0	6	1.5576	0.7273
No log	7.0	7	1.6687	0.7273
No log	8.0	8	1.5015	0.7273
No log	9.0	9	1.2729	0.7273
0.3584	10.0	10	1.2127	0.7273
0.3584	11.0	11	1.3132	0.7273
0.3584	12.0	12	1.3350	0.7273
0.3584	13.0	13	1.2579	0.7273
0.3584	14.0	14	1.3559	0.7273
0.3584	15.0	15	1.4231	0.7273
0.3584	16.0	16	1.5141	0.7273
0.3584	17.0	17	1.4200	0.7273
0.3584	18.0	18	1.2498	0.7273
0.3584	19.0	19	1.1456	0.7273
0.1919	20.0	20	1.1055	0.7273
0.1919	21.0	21	1.1937	0.7273
0.1919	22.0	22	1.2768	0.7273
0.1919	23.0	23	1.3224	0.7273
0.1919	24.0	24	1.3629	0.7273
0.1919	25.0	25	1.3238	0.7273
0.1919	26.0	26	1.2280	0.7273
0.1919	27.0	27	1.2446	0.7273
0.1919	28.0	28	1.2530	0.7273
0.1919	29.0	29	1.2468	0.7273
0.1447	30.0	30	1.1535	0.7273
0.1447	31.0	31	1.1125	0.7273
0.1447	32.0	32	1.2051	0.7273
0.1447	33.0	33	1.5902	0.7273
0.1447	34.0	34	1.8445	0.7273
0.1447	35.0	35	1.7222	0.7273
0.1447	36.0	36	1.5080	0.7273
0.1447	37.0	37	1.3542	0.7273
0.1447	38.0	38	1.3106	0.7273
0.1447	39.0	39	1.4533	0.7273
0.1053	40.0	40	1.6427	0.7273
0.1053	41.0	41	1.7518	0.7273
0.1053	42.0	42	1.7775	0.7273
0.1053	43.0	43	1.6831	0.7273
0.1053	44.0	44	1.6968	0.7273
0.1053	45.0	45	1.8236	0.7273
0.1053	46.0	46	1.8845	0.7273
0.1053	47.0	47	1.8785	0.7273
0.1053	48.0	48	1.8805	0.7273
0.1053	49.0	49	1.9625	0.7273
0.0771	50.0	50	1.9860	0.7273
0.0771	51.0	51	1.9708	0.7273
0.0771	52.0	52	1.9149	0.7273
0.0771	53.0	53	1.9064	0.7273
0.0771	54.0	54	1.8804	0.7273
0.0771	55.0	55	1.8467	0.7273
0.0771	56.0	56	1.8508	0.7273
0.0771	57.0	57	1.8675	0.7273
0.0771	58.0	58	1.8886	0.7273
0.0771	59.0	59	1.8860	0.7273
0.0528	60.0	60	1.8777	0.7273
0.0528	61.0	61	1.9119	0.7273
0.0528	62.0	62	1.9860	0.7273
0.0528	63.0	63	2.1003	0.7273
0.0528	64.0	64	2.1561	0.7273
0.0528	65.0	65	2.1454	0.7273
0.0528	66.0	66	2.0685	0.7273
0.0528	67.0	67	1.9261	0.7273
0.0528	68.0	68	1.6839	0.7273
0.0528	69.0	69	1.4306	0.7273
0.043	70.0	70	1.3800	0.7273
0.043	71.0	71	1.4814	0.7273
0.043	72.0	72	1.6014	0.7273
0.043	73.0	73	1.7792	0.7273
0.043	74.0	74	1.9423	0.7273
0.043	75.0	75	2.0590	0.7273
0.043	76.0	76	2.1119	0.7273
0.043	77.0	77	2.1116	0.7273
0.043	78.0	78	2.0979	0.7273
0.043	79.0	79	2.1457	0.7273
0.0429	80.0	80	2.2222	0.7273
0.0429	81.0	81	2.2803	0.7273
0.0429	82.0	82	2.3327	0.7273
0.0429	83.0	83	2.3643	0.7273
0.0429	84.0	84	2.3774	0.7273
0.0429	85.0	85	2.3838	0.7273
0.0429	86.0	86	2.4072	0.7273
0.0429	87.0	87	2.4189	0.7273
0.0429	88.0	88	2.4028	0.7273
0.0429	89.0	89	2.3847	0.7273
0.0475	90.0	90	2.3793	0.7273
0.0475	91.0	91	2.3831	0.7273
0.0475	92.0	92	2.3882	0.7273
0.0475	93.0	93	2.3964	0.7273
0.0475	94.0	94	2.4126	0.7273
0.0475	95.0	95	2.4309	0.7273
0.0475	96.0	96	2.4486	0.7273
0.0475	97.0	97	2.4628	0.7273
0.0475	98.0	98	2.4723	0.7273
0.0475	99.0	99	2.4775	0.7273
0.0337	100.0	100	2.4788	0.7273

Framework versions

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