beit-base-patch16-224-hasta-75-fold1

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: 0.2432
• 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	2.2077	0.0
No log	2.0	2	1.7828	0.0
No log	3.0	3	1.0543	0.3333
No log	4.0	4	0.4305	0.9167
No log	5.0	5	0.2672	0.9167
No log	6.0	6	0.3206	0.9167
No log	7.0	7	0.3318	0.9167
No log	8.0	8	0.3007	0.9167
No log	9.0	9	0.3268	0.9167
0.4863	10.0	10	0.3425	0.9167
0.4863	11.0	11	0.2906	0.9167
0.4863	12.0	12	0.2639	0.9167
0.4863	13.0	13	0.2962	0.9167
0.4863	14.0	14	0.4442	0.8333
0.4863	15.0	15	0.3108	0.8333
0.4863	16.0	16	0.2321	0.9167
0.4863	17.0	17	0.2309	0.9167
0.4863	18.0	18	0.2432	1.0
0.4863	19.0	19	0.2240	1.0
0.1603	20.0	20	0.1608	0.9167
0.1603	21.0	21	0.1275	0.9167
0.1603	22.0	22	0.1191	0.9167
0.1603	23.0	23	0.1030	0.9167
0.1603	24.0	24	0.1010	1.0
0.1603	25.0	25	0.0816	1.0
0.1603	26.0	26	0.1814	0.9167
0.1603	27.0	27	0.1654	0.9167
0.1603	28.0	28	0.0945	1.0
0.1603	29.0	29	0.0847	1.0
0.1007	30.0	30	0.1566	1.0
0.1007	31.0	31	0.0819	1.0
0.1007	32.0	32	0.0782	1.0
0.1007	33.0	33	0.0781	1.0
0.1007	34.0	34	0.0635	1.0
0.1007	35.0	35	0.0675	1.0
0.1007	36.0	36	0.1137	1.0
0.1007	37.0	37	0.1267	0.9167
0.1007	38.0	38	0.1438	0.9167
0.1007	39.0	39	0.1301	0.9167
0.0573	40.0	40	0.1123	0.9167
0.0573	41.0	41	0.0673	1.0
0.0573	42.0	42	0.0265	1.0
0.0573	43.0	43	0.0317	1.0
0.0573	44.0	44	0.0461	1.0
0.0573	45.0	45	0.0326	1.0
0.0573	46.0	46	0.0221	1.0
0.0573	47.0	47	0.0227	1.0
0.0573	48.0	48	0.0214	1.0
0.0573	49.0	49	0.0176	1.0
0.0566	50.0	50	0.0150	1.0
0.0566	51.0	51	0.0154	1.0
0.0566	52.0	52	0.0139	1.0
0.0566	53.0	53	0.0097	1.0
0.0566	54.0	54	0.0143	1.0
0.0566	55.0	55	0.0272	1.0
0.0566	56.0	56	0.0427	1.0
0.0566	57.0	57	0.0343	1.0
0.0566	58.0	58	0.0290	1.0
0.0566	59.0	59	0.0557	1.0
0.0242	60.0	60	0.0905	1.0
0.0242	61.0	61	0.1374	0.9167
0.0242	62.0	62	0.1763	0.9167
0.0242	63.0	63	0.1793	0.9167
0.0242	64.0	64	0.1640	0.9167
0.0242	65.0	65	0.1445	0.9167
0.0242	66.0	66	0.1092	1.0
0.0242	67.0	67	0.0915	1.0
0.0242	68.0	68	0.0640	1.0
0.0242	69.0	69	0.0376	1.0
0.0339	70.0	70	0.0297	1.0
0.0339	71.0	71	0.0238	1.0
0.0339	72.0	72	0.0178	1.0
0.0339	73.0	73	0.0104	1.0
0.0339	74.0	74	0.0063	1.0
0.0339	75.0	75	0.0042	1.0
0.0339	76.0	76	0.0031	1.0
0.0339	77.0	77	0.0029	1.0
0.0339	78.0	78	0.0034	1.0
0.0339	79.0	79	0.0035	1.0
0.0532	80.0	80	0.0035	1.0
0.0532	81.0	81	0.0039	1.0
0.0532	82.0	82	0.0054	1.0
0.0532	83.0	83	0.0110	1.0
0.0532	84.0	84	0.0255	1.0
0.0532	85.0	85	0.0500	1.0
0.0532	86.0	86	0.0844	0.9167
0.0532	87.0	87	0.1191	0.9167
0.0532	88.0	88	0.1437	0.9167
0.0532	89.0	89	0.1564	0.9167
0.0316	90.0	90	0.1544	0.9167
0.0316	91.0	91	0.1455	0.9167
0.0316	92.0	92	0.1383	0.9167
0.0316	93.0	93	0.1194	0.9167
0.0316	94.0	94	0.1027	0.9167
0.0316	95.0	95	0.0875	0.9167
0.0316	96.0	96	0.0715	1.0
0.0316	97.0	97	0.0608	1.0
0.0316	98.0	98	0.0519	1.0
0.0316	99.0	99	0.0468	1.0
0.0299	100.0	100	0.0442	1.0

Framework versions

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