smids_3x_deit_small_adamax_00001_fold3

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

• Loss: 0.7023
• Accuracy: 0.91

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
0.4073	1.0	225	0.3484	0.8667
0.2568	2.0	450	0.2863	0.9
0.1796	3.0	675	0.2552	0.9083
0.1374	4.0	900	0.2541	0.9067
0.1027	5.0	1125	0.2543	0.91
0.0731	6.0	1350	0.2688	0.905
0.0643	7.0	1575	0.2928	0.9117
0.0439	8.0	1800	0.3138	0.9133
0.0402	9.0	2025	0.3569	0.9083
0.0289	10.0	2250	0.4503	0.905
0.0259	11.0	2475	0.4729	0.9033
0.009	12.0	2700	0.4822	0.9117
0.0012	13.0	2925	0.5123	0.91
0.0039	14.0	3150	0.5071	0.905
0.0007	15.0	3375	0.5292	0.9083
0.0062	16.0	3600	0.5423	0.91
0.0007	17.0	3825	0.6134	0.9117
0.0002	18.0	4050	0.5764	0.9083
0.0001	19.0	4275	0.5936	0.91
0.0001	20.0	4500	0.5995	0.9117
0.0001	21.0	4725	0.6209	0.91
0.0001	22.0	4950	0.6160	0.915
0.0	23.0	5175	0.6182	0.9117
0.0	24.0	5400	0.6761	0.9067
0.0027	25.0	5625	0.6972	0.9083
0.0	26.0	5850	0.6567	0.9117
0.0	27.0	6075	0.6572	0.9133
0.0	28.0	6300	0.6611	0.9133
0.0018	29.0	6525	0.7006	0.9133
0.0016	30.0	6750	0.6678	0.9117
0.0133	31.0	6975	0.6927	0.9117
0.0	32.0	7200	0.6752	0.9117
0.0	33.0	7425	0.6754	0.915
0.0	34.0	7650	0.6853	0.9117
0.0	35.0	7875	0.6889	0.91
0.0	36.0	8100	0.6852	0.9117
0.0	37.0	8325	0.7018	0.9117
0.0	38.0	8550	0.6889	0.9133
0.0	39.0	8775	0.6896	0.9133
0.0	40.0	9000	0.6924	0.9133
0.0	41.0	9225	0.7070	0.9117
0.0	42.0	9450	0.6982	0.91
0.0	43.0	9675	0.6970	0.9133
0.0	44.0	9900	0.6943	0.9117
0.0	45.0	10125	0.7012	0.91
0.0	46.0	10350	0.7058	0.91
0.0076	47.0	10575	0.7019	0.91
0.0	48.0	10800	0.7041	0.91
0.0	49.0	11025	0.7038	0.91
0.0	50.0	11250	0.7023	0.91

Framework versions

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