smids_5x_beit_base_sgd_001_fold5

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: 0.2832
• Accuracy: 0.8867

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: 0.001
• 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
0.6263	1.0	375	0.6777	0.725
0.5866	2.0	750	0.4870	0.81
0.4483	3.0	1125	0.4247	0.825
0.4271	4.0	1500	0.3855	0.84
0.4066	5.0	1875	0.3633	0.8467
0.3828	6.0	2250	0.3474	0.8417
0.309	7.0	2625	0.3371	0.8583
0.3188	8.0	3000	0.3295	0.86
0.3147	9.0	3375	0.3210	0.8633
0.2842	10.0	3750	0.3163	0.8633
0.258	11.0	4125	0.3059	0.87
0.2796	12.0	4500	0.3036	0.8717
0.2552	13.0	4875	0.2994	0.87
0.2763	14.0	5250	0.2979	0.8633
0.2925	15.0	5625	0.3004	0.865
0.2222	16.0	6000	0.2915	0.8767
0.2839	17.0	6375	0.2879	0.8783
0.2546	18.0	6750	0.2876	0.88
0.2528	19.0	7125	0.2899	0.8817
0.1895	20.0	7500	0.2841	0.885
0.2366	21.0	7875	0.2901	0.8767
0.2149	22.0	8250	0.2831	0.8883
0.2987	23.0	8625	0.2845	0.8833
0.232	24.0	9000	0.2818	0.885
0.2416	25.0	9375	0.2809	0.8883
0.2147	26.0	9750	0.2789	0.8867
0.2824	27.0	10125	0.2796	0.8883
0.2229	28.0	10500	0.2814	0.8883
0.2625	29.0	10875	0.2884	0.8767
0.1908	30.0	11250	0.2826	0.885
0.2464	31.0	11625	0.2786	0.8867
0.2333	32.0	12000	0.2809	0.89
0.2568	33.0	12375	0.2768	0.8867
0.2444	34.0	12750	0.2777	0.8883
0.1971	35.0	13125	0.2787	0.8883
0.1586	36.0	13500	0.2808	0.8867
0.1628	37.0	13875	0.2838	0.8817
0.2206	38.0	14250	0.2772	0.8867
0.1707	39.0	14625	0.2818	0.8833
0.2328	40.0	15000	0.2820	0.8867
0.1705	41.0	15375	0.2828	0.89
0.1753	42.0	15750	0.2851	0.8867
0.2269	43.0	16125	0.2832	0.8933
0.1772	44.0	16500	0.2830	0.8883
0.235	45.0	16875	0.2841	0.8883
0.251	46.0	17250	0.2828	0.8867
0.2199	47.0	17625	0.2831	0.8883
0.1679	48.0	18000	0.2835	0.8867
0.2096	49.0	18375	0.2833	0.8867
0.22	50.0	18750	0.2832	0.8867

Framework versions

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