smids_3x_deit_base_adamax_00001_fold2

This model is a fine-tuned version of facebook/deit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.7483
• Accuracy: 0.8819

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
0.2908	1.0	225	0.3519	0.8469
0.2412	2.0	450	0.3425	0.8652
0.1561	3.0	675	0.3113	0.8752
0.1722	4.0	900	0.3333	0.8819
0.0793	5.0	1125	0.3397	0.8869
0.0512	6.0	1350	0.3703	0.8902
0.0408	7.0	1575	0.3948	0.8869
0.0153	8.0	1800	0.4474	0.8852
0.0036	9.0	2025	0.5055	0.8819
0.0009	10.0	2250	0.5138	0.8918
0.0011	11.0	2475	0.5776	0.8752
0.0004	12.0	2700	0.6002	0.8785
0.0004	13.0	2925	0.6053	0.8819
0.0002	14.0	3150	0.6097	0.8918
0.0002	15.0	3375	0.6366	0.8819
0.0001	16.0	3600	0.6507	0.8819
0.0042	17.0	3825	0.6732	0.8869
0.0001	18.0	4050	0.6626	0.8852
0.0001	19.0	4275	0.6800	0.8885
0.0001	20.0	4500	0.6886	0.8852
0.0001	21.0	4725	0.7001	0.8819
0.0001	22.0	4950	0.7256	0.8869
0.008	23.0	5175	0.7472	0.8918
0.0001	24.0	5400	0.7160	0.8835
0.0075	25.0	5625	0.7354	0.8852
0.0001	26.0	5850	0.7213	0.8819
0.0001	27.0	6075	0.7101	0.8835
0.0	28.0	6300	0.7245	0.8819
0.0001	29.0	6525	0.7475	0.8869
0.0058	30.0	6750	0.7235	0.8852
0.0	31.0	6975	0.7151	0.8852
0.0	32.0	7200	0.7303	0.8852
0.0	33.0	7425	0.7353	0.8835
0.0	34.0	7650	0.7337	0.8835
0.0	35.0	7875	0.7550	0.8835
0.0034	36.0	8100	0.7409	0.8885
0.0	37.0	8325	0.7323	0.8769
0.0	38.0	8550	0.7381	0.8835
0.0026	39.0	8775	0.7392	0.8819
0.0	40.0	9000	0.7428	0.8819
0.0	41.0	9225	0.7496	0.8835
0.0	42.0	9450	0.7433	0.8835
0.0	43.0	9675	0.7393	0.8852
0.0	44.0	9900	0.7435	0.8835
0.0	45.0	10125	0.7479	0.8835
0.0	46.0	10350	0.7454	0.8835
0.0	47.0	10575	0.7458	0.8802
0.0	48.0	10800	0.7464	0.8819
0.0037	49.0	11025	0.7477	0.8835
0.0037	50.0	11250	0.7483	0.8819

Framework versions

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