smids_3x_beit_base_adamax_0001_fold2

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: 0.9159
• Accuracy: 0.8902

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
0.3022	1.0	225	0.2725	0.8935
0.178	2.0	450	0.3008	0.8852
0.0845	3.0	675	0.4512	0.8686
0.0684	4.0	900	0.3867	0.8852
0.0813	5.0	1125	0.6871	0.8686
0.046	6.0	1350	0.5810	0.8802
0.0277	7.0	1575	0.6980	0.8769
0.0479	8.0	1800	0.6091	0.9002
0.0027	9.0	2025	0.7134	0.8952
0.0066	10.0	2250	0.6880	0.8952
0.0321	11.0	2475	0.6460	0.8902
0.0014	12.0	2700	0.7817	0.8935
0.0002	13.0	2925	0.6934	0.8935
0.021	14.0	3150	0.7639	0.9002
0.0064	15.0	3375	0.7684	0.8918
0.0006	16.0	3600	0.8221	0.8802
0.0143	17.0	3825	0.7209	0.8902
0.0001	18.0	4050	0.6982	0.8918
0.0001	19.0	4275	0.7862	0.9002
0.0027	20.0	4500	0.7966	0.8819
0.0001	21.0	4725	0.8116	0.8918
0.007	22.0	4950	0.9903	0.8869
0.0034	23.0	5175	0.8839	0.8935
0.0	24.0	5400	0.8613	0.8885
0.005	25.0	5625	0.8407	0.8935
0.0001	26.0	5850	0.8776	0.8968
0.0	27.0	6075	0.8976	0.8885
0.0	28.0	6300	0.8439	0.8918
0.0002	29.0	6525	0.8561	0.9035
0.0034	30.0	6750	0.8784	0.9035
0.0001	31.0	6975	1.0043	0.8835
0.0	32.0	7200	0.9310	0.8968
0.0001	33.0	7425	0.9435	0.8985
0.0002	34.0	7650	0.9440	0.8885
0.0	35.0	7875	0.9680	0.8869
0.0033	36.0	8100	0.9425	0.8952
0.0	37.0	8325	0.8592	0.9018
0.0001	38.0	8550	0.8640	0.8985
0.0029	39.0	8775	0.9094	0.8935
0.0	40.0	9000	0.8962	0.8968
0.0	41.0	9225	0.9188	0.8935
0.0	42.0	9450	0.9151	0.8968
0.0	43.0	9675	0.9033	0.9002
0.0	44.0	9900	0.9139	0.8968
0.0004	45.0	10125	0.9130	0.8968
0.0	46.0	10350	0.9128	0.8985
0.0	47.0	10575	0.9094	0.8952
0.0	48.0	10800	0.9134	0.8952
0.0032	49.0	11025	0.9151	0.8902
0.0002	50.0	11250	0.9159	0.8902

Framework versions

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