beit-base-patch16-224-hasta-75-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: 0.0761
• Accuracy: 1.0

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	0.9045	0.3333
No log	2.0	2	0.6690	0.8333
No log	3.0	3	0.4176	0.9167
No log	4.0	4	0.3611	0.9167
No log	5.0	5	0.3914	0.9167
No log	6.0	6	0.3954	0.9167
No log	7.0	7	0.3574	0.9167
No log	8.0	8	0.3282	0.9167
No log	9.0	9	0.3443	0.9167
0.3098	10.0	10	0.2939	0.9167
0.3098	11.0	11	0.2509	0.9167
0.3098	12.0	12	0.2714	0.9167
0.3098	13.0	13	0.3388	0.9167
0.3098	14.0	14	0.4240	0.8333
0.3098	15.0	15	0.3483	0.8333
0.3098	16.0	16	0.3874	0.75
0.3098	17.0	17	0.3755	0.8333
0.3098	18.0	18	0.2999	0.9167
0.3098	19.0	19	0.3344	0.9167
0.1617	20.0	20	0.3177	0.9167
0.1617	21.0	21	0.3033	0.8333
0.1617	22.0	22	0.2805	0.8333
0.1617	23.0	23	0.2428	0.8333
0.1617	24.0	24	0.1912	0.9167
0.1617	25.0	25	0.1992	0.8333
0.1617	26.0	26	0.2689	0.9167
0.1617	27.0	27	0.2284	0.9167
0.1617	28.0	28	0.1536	0.9167
0.1617	29.0	29	0.1593	0.9167
0.1003	30.0	30	0.1818	0.8333
0.1003	31.0	31	0.2490	0.8333
0.1003	32.0	32	0.3354	0.9167
0.1003	33.0	33	0.3148	0.8333
0.1003	34.0	34	0.3323	0.8333
0.1003	35.0	35	0.3582	0.9167
0.1003	36.0	36	0.3736	0.8333
0.1003	37.0	37	0.4285	0.8333
0.1003	38.0	38	0.4383	0.8333
0.1003	39.0	39	0.4500	0.8333
0.0541	40.0	40	0.3576	0.8333
0.0541	41.0	41	0.2811	0.8333
0.0541	42.0	42	0.1908	0.9167
0.0541	43.0	43	0.1601	0.9167
0.0541	44.0	44	0.1516	0.9167
0.0541	45.0	45	0.0944	0.9167
0.0541	46.0	46	0.1231	0.9167
0.0541	47.0	47	0.1886	0.9167
0.0541	48.0	48	0.1905	0.9167
0.0541	49.0	49	0.2160	0.9167
0.0479	50.0	50	0.1523	0.9167
0.0479	51.0	51	0.0761	1.0
0.0479	52.0	52	0.0611	1.0
0.0479	53.0	53	0.0579	1.0
0.0479	54.0	54	0.0918	0.9167
0.0479	55.0	55	0.2574	0.9167
0.0479	56.0	56	0.4721	0.9167
0.0479	57.0	57	0.5495	0.9167
0.0479	58.0	58	0.5856	0.9167
0.0479	59.0	59	0.5852	0.9167
0.0579	60.0	60	0.5607	0.9167
0.0579	61.0	61	0.4982	0.9167
0.0579	62.0	62	0.4343	0.9167
0.0579	63.0	63	0.3539	0.9167
0.0579	64.0	64	0.2492	0.9167
0.0579	65.0	65	0.2232	0.9167
0.0579	66.0	66	0.2359	0.9167
0.0579	67.0	67	0.2633	0.9167
0.0579	68.0	68	0.3047	0.9167
0.0579	69.0	69	0.3410	0.9167
0.0396	70.0	70	0.3630	0.9167
0.0396	71.0	71	0.3515	0.9167
0.0396	72.0	72	0.3256	0.9167
0.0396	73.0	73	0.2986	0.9167
0.0396	74.0	74	0.2439	0.9167
0.0396	75.0	75	0.1829	0.9167
0.0396	76.0	76	0.1434	0.9167
0.0396	77.0	77	0.1233	0.9167
0.0396	78.0	78	0.1253	0.9167
0.0396	79.0	79	0.1324	0.9167
0.0276	80.0	80	0.1492	0.9167
0.0276	81.0	81	0.1580	0.9167
0.0276	82.0	82	0.1729	0.9167
0.0276	83.0	83	0.1670	0.9167
0.0276	84.0	84	0.1609	0.9167
0.0276	85.0	85	0.1458	0.9167
0.0276	86.0	86	0.1262	0.9167
0.0276	87.0	87	0.1097	0.9167
0.0276	88.0	88	0.0934	0.9167
0.0276	89.0	89	0.0820	0.9167
0.0324	90.0	90	0.0731	1.0
0.0324	91.0	91	0.0676	1.0
0.0324	92.0	92	0.0651	1.0
0.0324	93.0	93	0.0627	1.0
0.0324	94.0	94	0.0610	1.0
0.0324	95.0	95	0.0600	1.0
0.0324	96.0	96	0.0592	1.0
0.0324	97.0	97	0.0586	1.0
0.0324	98.0	98	0.0579	1.0
0.0324	99.0	99	0.0572	1.0
0.0361	100.0	100	0.0569	1.0

Framework versions

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