smids_5x_beit_base_sgd_001_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: 0.2734
• Accuracy: 0.91

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
0.6727	1.0	375	0.6864	0.7317
0.6041	2.0	750	0.5073	0.8033
0.4189	3.0	1125	0.4416	0.8267
0.3943	4.0	1500	0.3997	0.8433
0.4289	5.0	1875	0.3785	0.855
0.3775	6.0	2250	0.3585	0.8617
0.3696	7.0	2625	0.3398	0.8767
0.3454	8.0	3000	0.3318	0.8767
0.3235	9.0	3375	0.3277	0.8717
0.3129	10.0	3750	0.3193	0.8733
0.2807	11.0	4125	0.3194	0.8833
0.2877	12.0	4500	0.3054	0.8867
0.2952	13.0	4875	0.3044	0.89
0.2764	14.0	5250	0.3000	0.89
0.2449	15.0	5625	0.2977	0.8933
0.2466	16.0	6000	0.2964	0.8983
0.2903	17.0	6375	0.2961	0.8883
0.2759	18.0	6750	0.2922	0.895
0.2465	19.0	7125	0.2899	0.9
0.2405	20.0	7500	0.2859	0.8967
0.1951	21.0	7875	0.2887	0.9
0.2066	22.0	8250	0.2843	0.9017
0.2962	23.0	8625	0.2838	0.9033
0.2403	24.0	9000	0.2814	0.9033
0.2279	25.0	9375	0.2878	0.9033
0.2748	26.0	9750	0.2833	0.9017
0.2265	27.0	10125	0.2865	0.905
0.2583	28.0	10500	0.2821	0.9083
0.1555	29.0	10875	0.2828	0.905
0.2329	30.0	11250	0.2785	0.9117
0.235	31.0	11625	0.2766	0.9067
0.245	32.0	12000	0.2760	0.91
0.2626	33.0	12375	0.2763	0.9067
0.2391	34.0	12750	0.2794	0.905
0.1618	35.0	13125	0.2760	0.9117
0.1815	36.0	13500	0.2756	0.9067
0.1967	37.0	13875	0.2796	0.9033
0.1639	38.0	14250	0.2740	0.91
0.2156	39.0	14625	0.2769	0.9083
0.1866	40.0	15000	0.2767	0.915
0.1684	41.0	15375	0.2758	0.9083
0.194	42.0	15750	0.2754	0.9067
0.2588	43.0	16125	0.2756	0.9083
0.213	44.0	16500	0.2746	0.91
0.1893	45.0	16875	0.2745	0.9083
0.2111	46.0	17250	0.2742	0.9083
0.2091	47.0	17625	0.2740	0.91
0.2023	48.0	18000	0.2739	0.9083
0.1852	49.0	18375	0.2735	0.91
0.197	50.0	18750	0.2734	0.91

Framework versions

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