smids_3x_deit_base_adamax_001_fold3

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: 0.9898
• Accuracy: 0.8967

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
0.4508	1.0	225	0.3728	0.8383
0.2776	2.0	450	0.4888	0.8267
0.2821	3.0	675	0.2991	0.8833
0.1623	4.0	900	0.3264	0.8783
0.1557	5.0	1125	0.3651	0.8833
0.1468	6.0	1350	0.4934	0.8667
0.0701	7.0	1575	0.6415	0.8583
0.1055	8.0	1800	0.4741	0.8717
0.0972	9.0	2025	0.4804	0.875
0.0586	10.0	2250	0.5016	0.8817
0.0818	11.0	2475	0.5231	0.8767
0.0651	12.0	2700	0.4690	0.89
0.0158	13.0	2925	0.6006	0.885
0.0389	14.0	3150	0.5509	0.8883
0.0218	15.0	3375	0.5621	0.88
0.0109	16.0	3600	0.5877	0.8933
0.0157	17.0	3825	0.5304	0.895
0.0158	18.0	4050	0.5697	0.895
0.0096	19.0	4275	0.6524	0.8983
0.0005	20.0	4500	0.6404	0.89
0.0001	21.0	4725	0.6926	0.8983
0.0045	22.0	4950	0.6417	0.8817
0.0215	23.0	5175	0.6424	0.89
0.0001	24.0	5400	0.7974	0.8683
0.0122	25.0	5625	0.7040	0.88
0.0	26.0	5850	0.7184	0.9083
0.0001	27.0	6075	0.6230	0.9067
0.0	28.0	6300	0.7043	0.9
0.0064	29.0	6525	0.7463	0.8983
0.0053	30.0	6750	0.7408	0.8983
0.0035	31.0	6975	0.7858	0.8983
0.0	32.0	7200	0.8160	0.9067
0.0	33.0	7425	0.8603	0.8967
0.0	34.0	7650	0.8311	0.9
0.0	35.0	7875	0.8519	0.905
0.0	36.0	8100	0.8622	0.8967
0.0032	37.0	8325	0.8530	0.8983
0.0	38.0	8550	0.9174	0.8967
0.0	39.0	8775	0.9290	0.9017
0.0	40.0	9000	0.9267	0.9017
0.0	41.0	9225	0.9188	0.8933
0.0	42.0	9450	0.9352	0.8933
0.0	43.0	9675	0.9338	0.9
0.0	44.0	9900	0.9386	0.9
0.0	45.0	10125	0.9613	0.8983
0.0	46.0	10350	0.9675	0.8933
0.0026	47.0	10575	0.9741	0.8967
0.0	48.0	10800	0.9808	0.8967
0.0	49.0	11025	0.9868	0.8967
0.0	50.0	11250	0.9898	0.8967

Framework versions

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