smids_1x_beit_base_adamax_0001_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: 0.6695
• Accuracy: 0.9133

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: 0.0001
• 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
0.3952	1.0	75	0.2900	0.885
0.2706	2.0	150	0.2858	0.895
0.1431	3.0	225	0.3509	0.8717
0.0811	4.0	300	0.2863	0.9183
0.0342	5.0	375	0.4874	0.8783
0.0334	6.0	450	0.4282	0.9117
0.0097	7.0	525	0.4216	0.92
0.0417	8.0	600	0.4392	0.91
0.0057	9.0	675	0.4243	0.9183
0.0023	10.0	750	0.5491	0.9
0.0342	11.0	825	0.4738	0.915
0.015	12.0	900	0.5105	0.9267
0.0179	13.0	975	0.6274	0.9083
0.0017	14.0	1050	0.5351	0.915
0.0012	15.0	1125	0.5446	0.905
0.0029	16.0	1200	0.5695	0.9067
0.0045	17.0	1275	0.5414	0.9133
0.0233	18.0	1350	0.7467	0.87
0.0001	19.0	1425	0.5934	0.9
0.0112	20.0	1500	0.5736	0.9067
0.0001	21.0	1575	0.6327	0.9033
0.0084	22.0	1650	0.5946	0.915
0.006	23.0	1725	0.5821	0.9133
0.0001	24.0	1800	0.6358	0.9
0.0	25.0	1875	0.5917	0.9117
0.0	26.0	1950	0.5998	0.9133
0.0002	27.0	2025	0.5967	0.915
0.0001	28.0	2100	0.5752	0.9117
0.0001	29.0	2175	0.6692	0.9
0.0044	30.0	2250	0.6493	0.9033
0.003	31.0	2325	0.6716	0.9117
0.0061	32.0	2400	0.7077	0.8983
0.0001	33.0	2475	0.6337	0.915
0.0003	34.0	2550	0.6698	0.9
0.0035	35.0	2625	0.6670	0.9033
0.0027	36.0	2700	0.6180	0.9067
0.0042	37.0	2775	0.6174	0.915
0.0023	38.0	2850	0.6161	0.9133
0.0001	39.0	2925	0.6601	0.91
0.0029	40.0	3000	0.6359	0.91
0.0	41.0	3075	0.6349	0.91
0.0022	42.0	3150	0.6576	0.9133
0.0028	43.0	3225	0.6662	0.9067
0.0	44.0	3300	0.6662	0.9083
0.0	45.0	3375	0.6797	0.9117
0.0	46.0	3450	0.6797	0.91
0.0043	47.0	3525	0.6738	0.91
0.0006	48.0	3600	0.6709	0.9133
0.0001	49.0	3675	0.6693	0.9133
0.0	50.0	3750	0.6695	0.9133

Framework versions

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