beit-base-patch16-224-hasta-75-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: 0.5330
• Accuracy: 0.9167

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	1.2661	0.1667
No log	2.0	2	0.9215	0.5
No log	3.0	3	0.5330	0.9167
No log	4.0	4	0.4361	0.9167
No log	5.0	5	0.4724	0.9167
No log	6.0	6	0.4424	0.9167
No log	7.0	7	0.3788	0.9167
No log	8.0	8	0.4228	0.9167
No log	9.0	9	0.4592	0.9167
0.3974	10.0	10	0.3966	0.9167
0.3974	11.0	11	0.3517	0.9167
0.3974	12.0	12	0.3481	0.9167
0.3974	13.0	13	0.3315	0.9167
0.3974	14.0	14	0.3353	0.9167
0.3974	15.0	15	0.3784	0.9167
0.3974	16.0	16	0.4232	0.9167
0.3974	17.0	17	0.4654	0.9167
0.3974	18.0	18	0.4091	0.9167
0.3974	19.0	19	0.4149	0.9167
0.1925	20.0	20	0.4202	0.9167
0.1925	21.0	21	0.4502	0.9167
0.1925	22.0	22	0.4371	0.9167
0.1925	23.0	23	0.4291	0.9167
0.1925	24.0	24	0.4265	0.8333
0.1925	25.0	25	0.4414	0.8333
0.1925	26.0	26	0.3957	0.8333
0.1925	27.0	27	0.3567	0.9167
0.1925	28.0	28	0.3406	0.9167
0.1925	29.0	29	0.3101	0.9167
0.1407	30.0	30	0.2956	0.9167
0.1407	31.0	31	0.3548	0.9167
0.1407	32.0	32	0.3067	0.9167
0.1407	33.0	33	0.2485	0.9167
0.1407	34.0	34	0.2818	0.9167
0.1407	35.0	35	0.3197	0.9167
0.1407	36.0	36	0.3401	0.9167
0.1407	37.0	37	0.3282	0.9167
0.1407	38.0	38	0.3078	0.9167
0.1407	39.0	39	0.2906	0.9167
0.1204	40.0	40	0.2875	0.9167
0.1204	41.0	41	0.3188	0.9167
0.1204	42.0	42	0.3449	0.9167
0.1204	43.0	43	0.3520	0.9167
0.1204	44.0	44	0.3401	0.9167
0.1204	45.0	45	0.3029	0.9167
0.1204	46.0	46	0.2584	0.9167
0.1204	47.0	47	0.2358	0.9167
0.1204	48.0	48	0.2265	0.9167
0.1204	49.0	49	0.2144	0.9167
0.0691	50.0	50	0.1622	0.9167
0.0691	51.0	51	0.1094	0.9167
0.0691	52.0	52	0.1955	0.9167
0.0691	53.0	53	0.3863	0.9167
0.0691	54.0	54	0.4803	0.9167
0.0691	55.0	55	0.5175	0.9167
0.0691	56.0	56	0.4899	0.9167
0.0691	57.0	57	0.4092	0.9167
0.0691	58.0	58	0.3755	0.9167
0.0691	59.0	59	0.3642	0.9167
0.062	60.0	60	0.4002	0.9167
0.062	61.0	61	0.4086	0.9167
0.062	62.0	62	0.4066	0.9167
0.062	63.0	63	0.3781	0.9167
0.062	64.0	64	0.3259	0.9167
0.062	65.0	65	0.2518	0.9167
0.062	66.0	66	0.2186	0.9167
0.062	67.0	67	0.2601	0.9167
0.062	68.0	68	0.2965	0.9167
0.062	69.0	69	0.3699	0.9167
0.0313	70.0	70	0.4417	0.9167
0.0313	71.0	71	0.5105	0.9167
0.0313	72.0	72	0.5439	0.9167
0.0313	73.0	73	0.5557	0.9167
0.0313	74.0	74	0.5514	0.9167
0.0313	75.0	75	0.5486	0.9167
0.0313	76.0	76	0.5317	0.9167
0.0313	77.0	77	0.4996	0.9167
0.0313	78.0	78	0.4638	0.9167
0.0313	79.0	79	0.4196	0.9167
0.0359	80.0	80	0.3639	0.9167
0.0359	81.0	81	0.3530	0.9167
0.0359	82.0	82	0.3918	0.9167
0.0359	83.0	83	0.4290	0.9167
0.0359	84.0	84	0.4569	0.9167
0.0359	85.0	85	0.4849	0.9167
0.0359	86.0	86	0.5136	0.9167
0.0359	87.0	87	0.5406	0.9167
0.0359	88.0	88	0.5586	0.9167
0.0359	89.0	89	0.5745	0.9167
0.0338	90.0	90	0.5878	0.9167
0.0338	91.0	91	0.5981	0.9167
0.0338	92.0	92	0.6071	0.9167
0.0338	93.0	93	0.6133	0.9167
0.0338	94.0	94	0.6139	0.9167
0.0338	95.0	95	0.6106	0.9167
0.0338	96.0	96	0.6059	0.9167
0.0338	97.0	97	0.6019	0.9167
0.0338	98.0	98	0.5981	0.9167
0.0338	99.0	99	0.5942	0.9167
0.0302	100.0	100	0.5923	0.9167

Framework versions

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