smids_3x_beit_base_rms_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: 1.6713
• Accuracy: 0.855

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
0.7637	1.0	225	0.7530	0.7217
0.5551	2.0	450	0.6161	0.7583
0.4831	3.0	675	0.4948	0.7833
0.3281	4.0	900	0.5414	0.8033
0.3506	5.0	1125	0.4226	0.815
0.3328	6.0	1350	0.4220	0.83
0.2581	7.0	1575	0.5786	0.7883
0.1949	8.0	1800	0.5329	0.8133
0.2071	9.0	2025	0.4652	0.8417
0.1906	10.0	2250	0.5303	0.82
0.1705	11.0	2475	0.6288	0.8283
0.153	12.0	2700	0.5236	0.8383
0.0688	13.0	2925	0.7459	0.8133
0.0899	14.0	3150	0.8275	0.8117
0.0904	15.0	3375	0.7966	0.8467
0.0822	16.0	3600	0.8714	0.835
0.112	17.0	3825	0.8906	0.8433
0.0635	18.0	4050	0.8797	0.835
0.0639	19.0	4275	0.8962	0.85
0.0335	20.0	4500	1.1500	0.815
0.0798	21.0	4725	0.9654	0.82
0.0508	22.0	4950	1.1138	0.8467
0.0254	23.0	5175	0.9088	0.8367
0.0398	24.0	5400	1.1000	0.83
0.083	25.0	5625	0.9695	0.8367
0.0354	26.0	5850	1.1614	0.8317
0.018	27.0	6075	1.1091	0.8533
0.0554	28.0	6300	1.0860	0.8417
0.0435	29.0	6525	1.0082	0.8533
0.0521	30.0	6750	1.0251	0.8333
0.0523	31.0	6975	1.1028	0.8267
0.0058	32.0	7200	1.2099	0.8433
0.0089	33.0	7425	1.4585	0.8383
0.0197	34.0	7650	1.2388	0.8483
0.0029	35.0	7875	1.3364	0.83
0.0016	36.0	8100	1.4458	0.8417
0.0092	37.0	8325	1.4004	0.835
0.0003	38.0	8550	1.4317	0.8417
0.0011	39.0	8775	1.2820	0.8417
0.0089	40.0	9000	1.5154	0.8417
0.0236	41.0	9225	1.3755	0.8467
0.0009	42.0	9450	1.6899	0.8517
0.0128	43.0	9675	1.5784	0.845
0.0006	44.0	9900	1.6022	0.8517
0.0002	45.0	10125	1.4557	0.8467
0.0206	46.0	10350	1.5017	0.855
0.006	47.0	10575	1.5387	0.855
0.0004	48.0	10800	1.6762	0.855
0.001	49.0	11025	1.7088	0.855
0.0003	50.0	11250	1.6713	0.855

Framework versions

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