smids_5x_deit_tiny_adamax_001_fold1

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: 0.9489
• Accuracy: 0.8898

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
0.3549	1.0	376	0.4844	0.8264
0.2678	2.0	752	0.3259	0.8798
0.3098	3.0	1128	0.3469	0.8548
0.2057	4.0	1504	0.3089	0.8831
0.15	5.0	1880	0.4280	0.8748
0.0947	6.0	2256	0.5773	0.8581
0.1544	7.0	2632	0.3805	0.8881
0.1085	8.0	3008	0.4878	0.8731
0.0399	9.0	3384	0.4495	0.8965
0.0251	10.0	3760	0.5573	0.8681
0.0684	11.0	4136	0.4467	0.8648
0.0506	12.0	4512	0.5126	0.8982
0.0075	13.0	4888	0.8575	0.8715
0.0481	14.0	5264	0.7463	0.8664
0.0077	15.0	5640	0.6816	0.8865
0.0098	16.0	6016	0.6312	0.8831
0.0003	17.0	6392	0.7022	0.8965
0.0075	18.0	6768	0.6976	0.8731
0.0042	19.0	7144	0.6012	0.8881
0.0311	20.0	7520	0.7693	0.8932
0.003	21.0	7896	0.6254	0.8915
0.0101	22.0	8272	0.6004	0.8998
0.0209	23.0	8648	0.7643	0.8815
0.0001	24.0	9024	0.8262	0.8848
0.0007	25.0	9400	0.6944	0.8898
0.0034	26.0	9776	0.7140	0.8915
0.0071	27.0	10152	0.8088	0.8798
0.0001	28.0	10528	0.7766	0.9032
0.0039	29.0	10904	0.8084	0.8948
0.0045	30.0	11280	0.7741	0.8831
0.0006	31.0	11656	0.8264	0.8932
0.0	32.0	12032	0.8432	0.8865
0.0	33.0	12408	0.8641	0.8848
0.0	34.0	12784	0.8447	0.8865
0.0	35.0	13160	0.8402	0.8848
0.0	36.0	13536	0.8232	0.8948
0.0	37.0	13912	0.8382	0.8915
0.0	38.0	14288	0.8652	0.8898
0.0	39.0	14664	0.8733	0.8848
0.0	40.0	15040	0.8254	0.8881
0.0	41.0	15416	0.8627	0.8848
0.0	42.0	15792	0.8799	0.8881
0.0	43.0	16168	0.8887	0.8915
0.0	44.0	16544	0.9046	0.8932
0.0	45.0	16920	0.9092	0.8932
0.0031	46.0	17296	0.9143	0.8881
0.0	47.0	17672	0.9293	0.8915
0.0	48.0	18048	0.9378	0.8898
0.0	49.0	18424	0.9447	0.8898
0.0023	50.0	18800	0.9489	0.8898

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.1+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2