smids_10x_beit_large_adamax_00001_fold4

This model is a fine-tuned version of microsoft/beit-large-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.1774
• Accuracy: 0.8933

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.174	1.0	750	0.3318	0.8817
0.0694	2.0	1500	0.3979	0.8833
0.0385	3.0	2250	0.6069	0.8817
0.0028	4.0	3000	0.7041	0.8767
0.0151	5.0	3750	0.8263	0.8783
0.017	6.0	4500	0.8468	0.8917
0.0004	7.0	5250	0.9156	0.8817
0.0149	8.0	6000	0.9947	0.8883
0.0019	9.0	6750	0.9986	0.8833
0.0	10.0	7500	1.0174	0.89
0.0002	11.0	8250	1.0347	0.8983
0.0006	12.0	9000	1.1212	0.8883
0.0007	13.0	9750	1.1145	0.9
0.002	14.0	10500	1.1511	0.895
0.0113	15.0	11250	1.1891	0.8833
0.0193	16.0	12000	1.1467	0.8833
0.0	17.0	12750	1.2067	0.8833
0.0	18.0	13500	1.1030	0.8917
0.0	19.0	14250	1.2269	0.8817
0.0	20.0	15000	1.2142	0.8983
0.0	21.0	15750	1.2333	0.8833
0.0	22.0	16500	1.2215	0.89
0.0	23.0	17250	1.1755	0.88
0.0001	24.0	18000	1.2025	0.89
0.0	25.0	18750	1.1234	0.8967
0.0	26.0	19500	1.1299	0.8933
0.0	27.0	20250	1.1278	0.8933
0.0	28.0	21000	1.1853	0.89
0.0	29.0	21750	1.1366	0.8967
0.0	30.0	22500	1.2109	0.8817
0.0	31.0	23250	1.2247	0.88
0.0124	32.0	24000	1.2057	0.885
0.0	33.0	24750	1.2082	0.8933
0.0	34.0	25500	1.1875	0.8933
0.0	35.0	26250	1.1823	0.8983
0.0	36.0	27000	1.1794	0.8883
0.0	37.0	27750	1.1760	0.8917
0.0	38.0	28500	1.1363	0.895
0.0	39.0	29250	1.1574	0.895
0.0	40.0	30000	1.1725	0.8933
0.0	41.0	30750	1.1844	0.8867
0.0	42.0	31500	1.1542	0.8933
0.0	43.0	32250	1.1472	0.895
0.0	44.0	33000	1.1640	0.8917
0.0	45.0	33750	1.1642	0.89
0.0	46.0	34500	1.1680	0.8933
0.0	47.0	35250	1.1880	0.895
0.0	48.0	36000	1.1744	0.8933
0.0	49.0	36750	1.1763	0.8933
0.0008	50.0	37500	1.1774	0.8933

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2