beit-base-patch16-224-hasta-85-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: 0.9074
• 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	1.4041	0.2727
No log	2.0	2	1.1167	0.5455
No log	3.0	3	0.7944	0.6364
No log	4.0	4	0.9074	0.7273
No log	5.0	5	1.3600	0.7273
No log	6.0	6	1.6643	0.7273
No log	7.0	7	1.6410	0.7273
No log	8.0	8	1.4549	0.7273
No log	9.0	9	1.4814	0.3636
0.4133	10.0	10	1.2719	0.6364
0.4133	11.0	11	1.0360	0.7273
0.4133	12.0	12	1.0691	0.7273
0.4133	13.0	13	0.9528	0.7273
0.4133	14.0	14	1.0430	0.7273
0.4133	15.0	15	1.3221	0.7273
0.4133	16.0	16	1.5236	0.7273
0.4133	17.0	17	1.5914	0.7273
0.4133	18.0	18	1.5015	0.7273
0.4133	19.0	19	1.3859	0.7273
0.1601	20.0	20	1.4514	0.7273
0.1601	21.0	21	1.6169	0.7273
0.1601	22.0	22	1.5223	0.7273
0.1601	23.0	23	1.4172	0.7273
0.1601	24.0	24	1.4753	0.6364
0.1601	25.0	25	1.5207	0.7273
0.1601	26.0	26	1.7055	0.7273
0.1601	27.0	27	1.7901	0.7273
0.1601	28.0	28	1.9109	0.7273
0.1601	29.0	29	1.8261	0.7273
0.1015	30.0	30	1.5657	0.7273
0.1015	31.0	31	1.3513	0.7273
0.1015	32.0	32	1.3374	0.7273
0.1015	33.0	33	1.4942	0.7273
0.1015	34.0	34	1.7731	0.7273
0.1015	35.0	35	1.8369	0.7273
0.1015	36.0	36	1.7328	0.7273
0.1015	37.0	37	1.5451	0.7273
0.1015	38.0	38	1.4483	0.7273
0.1015	39.0	39	1.4289	0.7273
0.0819	40.0	40	1.3922	0.7273
0.0819	41.0	41	1.4002	0.7273
0.0819	42.0	42	1.3673	0.7273
0.0819	43.0	43	1.2194	0.7273
0.0819	44.0	44	1.2879	0.7273
0.0819	45.0	45	1.4545	0.7273
0.0819	46.0	46	1.6902	0.7273
0.0819	47.0	47	1.8415	0.7273
0.0819	48.0	48	1.8891	0.7273
0.0819	49.0	49	1.8339	0.7273
0.0551	50.0	50	1.8433	0.7273
0.0551	51.0	51	1.8432	0.7273
0.0551	52.0	52	1.8002	0.7273
0.0551	53.0	53	1.7276	0.7273
0.0551	54.0	54	1.5764	0.7273
0.0551	55.0	55	1.3951	0.7273
0.0551	56.0	56	1.4243	0.7273
0.0551	57.0	57	1.5620	0.7273
0.0551	58.0	58	1.6897	0.7273
0.0551	59.0	59	1.7995	0.7273
0.0413	60.0	60	1.8739	0.7273
0.0413	61.0	61	1.8486	0.7273
0.0413	62.0	62	1.8073	0.7273
0.0413	63.0	63	1.8193	0.7273
0.0413	64.0	64	1.7604	0.7273
0.0413	65.0	65	1.6327	0.7273
0.0413	66.0	66	1.5447	0.7273
0.0413	67.0	67	1.4243	0.7273
0.0413	68.0	68	1.3810	0.7273
0.0413	69.0	69	1.3641	0.7273
0.038	70.0	70	1.4688	0.7273
0.038	71.0	71	1.5677	0.7273
0.038	72.0	72	1.7174	0.7273
0.038	73.0	73	1.7920	0.7273
0.038	74.0	74	1.9000	0.7273
0.038	75.0	75	1.9468	0.7273
0.038	76.0	76	1.9872	0.7273
0.038	77.0	77	2.0208	0.7273
0.038	78.0	78	2.0135	0.7273
0.038	79.0	79	1.9762	0.7273
0.0365	80.0	80	1.9576	0.7273
0.0365	81.0	81	1.9310	0.7273
0.0365	82.0	82	1.8495	0.7273
0.0365	83.0	83	1.7683	0.7273
0.0365	84.0	84	1.7109	0.7273
0.0365	85.0	85	1.6438	0.7273
0.0365	86.0	86	1.6154	0.7273
0.0365	87.0	87	1.5715	0.7273
0.0365	88.0	88	1.5428	0.7273
0.0365	89.0	89	1.5164	0.7273
0.038	90.0	90	1.5008	0.7273
0.038	91.0	91	1.4730	0.7273
0.038	92.0	92	1.4493	0.7273
0.038	93.0	93	1.4728	0.7273
0.038	94.0	94	1.5033	0.7273
0.038	95.0	95	1.5346	0.7273
0.038	96.0	96	1.5556	0.7273
0.038	97.0	97	1.5643	0.7273
0.038	98.0	98	1.5759	0.7273
0.038	99.0	99	1.5807	0.7273
0.0191	100.0	100	1.5806	0.7273

Framework versions

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