beit-base-patch16-224-hasta-75-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: 0.0769
• 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	0.5149	0.9167
No log	2.0	2	0.4041	0.9167
No log	3.0	3	0.3602	0.9167
No log	4.0	4	0.3914	0.9167
No log	5.0	5	0.4230	0.9167
No log	6.0	6	0.4505	0.9167
No log	7.0	7	0.4577	0.9167
No log	8.0	8	0.4562	0.9167
No log	9.0	9	0.4353	0.9167
0.2449	10.0	10	0.4424	0.9167
0.2449	11.0	11	0.4563	0.9167
0.2449	12.0	12	0.4720	0.8333
0.2449	13.0	13	0.3685	0.9167
0.2449	14.0	14	0.3112	0.9167
0.2449	15.0	15	0.2517	0.9167
0.2449	16.0	16	0.2199	0.9167
0.2449	17.0	17	0.2582	0.9167
0.2449	18.0	18	0.3222	0.9167
0.2449	19.0	19	0.2529	0.9167
0.1293	20.0	20	0.1724	0.9167
0.1293	21.0	21	0.1393	0.9167
0.1293	22.0	22	0.0965	0.9167
0.1293	23.0	23	0.0769	1.0
0.1293	24.0	24	0.0690	1.0
0.1293	25.0	25	0.1172	0.9167
0.1293	26.0	26	0.1629	0.9167
0.1293	27.0	27	0.1103	0.9167
0.1293	28.0	28	0.1193	1.0
0.1293	29.0	29	0.1003	1.0
0.0785	30.0	30	0.0719	1.0
0.0785	31.0	31	0.1636	0.9167
0.0785	32.0	32	0.2265	0.9167
0.0785	33.0	33	0.2009	0.9167
0.0785	34.0	34	0.1122	0.9167
0.0785	35.0	35	0.0424	1.0
0.0785	36.0	36	0.0407	1.0
0.0785	37.0	37	0.0322	1.0
0.0785	38.0	38	0.0397	1.0
0.0785	39.0	39	0.0569	1.0
0.0519	40.0	40	0.0891	0.9167
0.0519	41.0	41	0.0974	0.9167
0.0519	42.0	42	0.0969	0.9167
0.0519	43.0	43	0.1268	0.9167
0.0519	44.0	44	0.1718	0.9167
0.0519	45.0	45	0.2412	0.9167
0.0519	46.0	46	0.2935	0.9167
0.0519	47.0	47	0.3791	0.9167
0.0519	48.0	48	0.4504	0.9167
0.0519	49.0	49	0.4738	0.9167
0.0542	50.0	50	0.4470	0.9167
0.0542	51.0	51	0.3819	0.9167
0.0542	52.0	52	0.3327	0.9167
0.0542	53.0	53	0.2756	0.9167
0.0542	54.0	54	0.2012	0.9167
0.0542	55.0	55	0.1548	0.9167
0.0542	56.0	56	0.1668	0.9167
0.0542	57.0	57	0.2124	0.9167
0.0542	58.0	58	0.2389	0.9167
0.0542	59.0	59	0.2337	0.9167
0.0223	60.0	60	0.2025	0.9167
0.0223	61.0	61	0.1676	0.9167
0.0223	62.0	62	0.1085	0.9167
0.0223	63.0	63	0.0652	0.9167
0.0223	64.0	64	0.0644	0.9167
0.0223	65.0	65	0.0544	1.0
0.0223	66.0	66	0.0577	1.0
0.0223	67.0	67	0.0799	0.9167
0.0223	68.0	68	0.1005	0.9167
0.0223	69.0	69	0.1069	0.9167
0.0255	70.0	70	0.1410	0.9167
0.0255	71.0	71	0.1748	0.9167
0.0255	72.0	72	0.2266	0.9167
0.0255	73.0	73	0.2680	0.9167
0.0255	74.0	74	0.2917	0.9167
0.0255	75.0	75	0.3008	0.9167
0.0255	76.0	76	0.2897	0.9167
0.0255	77.0	77	0.2652	0.9167
0.0255	78.0	78	0.2542	0.9167
0.0255	79.0	79	0.2405	0.9167
0.0252	80.0	80	0.2391	0.9167
0.0252	81.0	81	0.2155	0.9167
0.0252	82.0	82	0.1953	0.9167
0.0252	83.0	83	0.1830	0.9167
0.0252	84.0	84	0.1672	0.9167
0.0252	85.0	85	0.1694	0.9167
0.0252	86.0	86	0.1743	0.9167
0.0252	87.0	87	0.1845	0.9167
0.0252	88.0	88	0.2098	0.9167
0.0252	89.0	89	0.2427	0.9167
0.0256	90.0	90	0.2682	0.9167
0.0256	91.0	91	0.2838	0.9167
0.0256	92.0	92	0.2936	0.9167
0.0256	93.0	93	0.3037	0.9167
0.0256	94.0	94	0.3065	0.9167
0.0256	95.0	95	0.3058	0.9167
0.0256	96.0	96	0.2997	0.9167
0.0256	97.0	97	0.2920	0.9167
0.0256	98.0	98	0.2849	0.9167
0.0256	99.0	99	0.2810	0.9167
0.0279	100.0	100	0.2798	0.9167

Framework versions

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