hushem_5x_beit_base_adamax_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: 0.2768
• Accuracy: 0.9286

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
1.2143	1.0	28	1.1034	0.5714
0.7775	2.0	56	0.7868	0.7143
0.4533	3.0	84	0.5700	0.8571
0.3037	4.0	112	0.4325	0.9048
0.1612	5.0	140	0.3176	0.9524
0.1058	6.0	168	0.2780	0.9524
0.0913	7.0	196	0.2215	0.9524
0.0514	8.0	224	0.1972	0.9524
0.0348	9.0	252	0.1996	0.9524
0.0351	10.0	280	0.1992	0.9524
0.0183	11.0	308	0.2116	0.9286
0.0237	12.0	336	0.2277	0.9286
0.0106	13.0	364	0.2270	0.9286
0.0108	14.0	392	0.2101	0.9524
0.0149	15.0	420	0.2231	0.9524
0.0076	16.0	448	0.2350	0.9048
0.0086	17.0	476	0.2204	0.9286
0.0033	18.0	504	0.2707	0.9286
0.0048	19.0	532	0.2227	0.9286
0.0041	20.0	560	0.2590	0.9286
0.0023	21.0	588	0.2904	0.9048
0.0045	22.0	616	0.2887	0.9286
0.0027	23.0	644	0.2955	0.9286
0.0033	24.0	672	0.2912	0.9286
0.0028	25.0	700	0.2636	0.9286
0.0018	26.0	728	0.2618	0.9286
0.0028	27.0	756	0.2893	0.9286
0.0019	28.0	784	0.2937	0.9286
0.0014	29.0	812	0.2912	0.9048
0.0031	30.0	840	0.2819	0.9048
0.0013	31.0	868	0.2819	0.9524
0.006	32.0	896	0.2996	0.9286
0.001	33.0	924	0.2836	0.9048
0.0011	34.0	952	0.2841	0.9286
0.0015	35.0	980	0.2638	0.9286
0.0022	36.0	1008	0.2845	0.9286
0.001	37.0	1036	0.2920	0.9286
0.0015	38.0	1064	0.2827	0.9286
0.0029	39.0	1092	0.2797	0.9286
0.002	40.0	1120	0.2954	0.9286
0.0017	41.0	1148	0.3039	0.9286
0.001	42.0	1176	0.3143	0.9286
0.0014	43.0	1204	0.3005	0.9286
0.0014	44.0	1232	0.2937	0.9286
0.0019	45.0	1260	0.2833	0.9286
0.0042	46.0	1288	0.2805	0.9286
0.0013	47.0	1316	0.2768	0.9286
0.0008	48.0	1344	0.2768	0.9286
0.0031	49.0	1372	0.2768	0.9286
0.0013	50.0	1400	0.2768	0.9286

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0