smids_5x_beit_base_rms_00001_fold4

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: 1.3109
• 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
0.1968	1.0	375	0.3151	0.8917
0.1474	2.0	750	0.4474	0.8667
0.0741	3.0	1125	0.4618	0.89
0.0638	4.0	1500	0.5553	0.9083
0.0357	5.0	1875	0.7199	0.8767
0.027	6.0	2250	0.8598	0.8783
0.0037	7.0	2625	1.0235	0.8817
0.0229	8.0	3000	1.0021	0.8817
0.0002	9.0	3375	1.0533	0.88
0.0003	10.0	3750	1.0170	0.8917
0.0047	11.0	4125	1.0274	0.885
0.0161	12.0	4500	0.9972	0.8883
0.0395	13.0	4875	1.1208	0.8817
0.0195	14.0	5250	1.1819	0.8833
0.0231	15.0	5625	1.2063	0.8867
0.002	16.0	6000	1.1906	0.8817
0.0189	17.0	6375	1.3367	0.8683
0.006	18.0	6750	1.3216	0.8767
0.0201	19.0	7125	1.2482	0.8883
0.0004	20.0	7500	1.3064	0.88
0.0	21.0	7875	1.2624	0.8833
0.0332	22.0	8250	1.2916	0.8783
0.0001	23.0	8625	1.2718	0.875
0.0134	24.0	9000	1.2861	0.8767
0.0091	25.0	9375	1.2558	0.8867
0.0	26.0	9750	1.1412	0.875
0.0003	27.0	10125	1.1757	0.8883
0.0	28.0	10500	1.1969	0.885
0.0001	29.0	10875	1.2159	0.8833
0.0439	30.0	11250	1.2112	0.885
0.0	31.0	11625	1.1996	0.8867
0.0011	32.0	12000	1.2726	0.8917
0.0	33.0	12375	1.2290	0.895
0.003	34.0	12750	1.2689	0.885
0.0001	35.0	13125	1.2685	0.8833
0.0	36.0	13500	1.2338	0.89
0.0	37.0	13875	1.2931	0.8817
0.0037	38.0	14250	1.2980	0.8833
0.0	39.0	14625	1.3078	0.895
0.0	40.0	15000	1.3295	0.8867
0.0	41.0	15375	1.2988	0.8917
0.0	42.0	15750	1.3679	0.8817
0.0	43.0	16125	1.3182	0.8833
0.0	44.0	16500	1.3785	0.885
0.0	45.0	16875	1.3130	0.8833
0.0008	46.0	17250	1.3226	0.8917
0.028	47.0	17625	1.3211	0.89
0.0214	48.0	18000	1.3040	0.8933
0.0	49.0	18375	1.3082	0.8917
0.0	50.0	18750	1.3109	0.8933

Framework versions

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