smids_5x_deit_tiny_adamax_00001_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: 0.9436
• Accuracy: 0.8831

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
0.3072	1.0	376	0.3156	0.8765
0.1866	2.0	752	0.3111	0.8798
0.1992	3.0	1128	0.2782	0.8881
0.1175	4.0	1504	0.2771	0.8932
0.0885	5.0	1880	0.3109	0.8948
0.0629	6.0	2256	0.3306	0.9015
0.0575	7.0	2632	0.3619	0.8948
0.0255	8.0	3008	0.4740	0.8881
0.0161	9.0	3384	0.5352	0.8765
0.0279	10.0	3760	0.5956	0.8815
0.0384	11.0	4136	0.6406	0.8831
0.0026	12.0	4512	0.7055	0.8748
0.0045	13.0	4888	0.7435	0.8715
0.0194	14.0	5264	0.7535	0.8798
0.0002	15.0	5640	0.8056	0.8748
0.0001	16.0	6016	0.8509	0.8781
0.0001	17.0	6392	0.8210	0.8765
0.0001	18.0	6768	0.8304	0.8781
0.0	19.0	7144	0.8626	0.8798
0.0304	20.0	7520	0.8719	0.8765
0.0	21.0	7896	0.9044	0.8781
0.0001	22.0	8272	0.8924	0.8881
0.0	23.0	8648	0.8973	0.8815
0.0	24.0	9024	0.8953	0.8798
0.0	25.0	9400	0.9328	0.8698
0.0041	26.0	9776	0.9144	0.8781
0.0	27.0	10152	0.9041	0.8798
0.0	28.0	10528	0.9154	0.8815
0.0022	29.0	10904	0.9165	0.8831
0.0086	30.0	11280	0.9177	0.8815
0.0	31.0	11656	0.9151	0.8815
0.0	32.0	12032	0.9187	0.8765
0.0	33.0	12408	0.9203	0.8815
0.0	34.0	12784	0.9308	0.8781
0.0	35.0	13160	0.9377	0.8781
0.0	36.0	13536	0.9294	0.8765
0.0	37.0	13912	0.9284	0.8831
0.0	38.0	14288	0.9287	0.8831
0.0	39.0	14664	0.9291	0.8831
0.0	40.0	15040	0.9319	0.8815
0.0	41.0	15416	0.9457	0.8781
0.0	42.0	15792	0.9438	0.8781
0.0	43.0	16168	0.9428	0.8798
0.0	44.0	16544	0.9398	0.8831
0.0	45.0	16920	0.9423	0.8831
0.0059	46.0	17296	0.9455	0.8781
0.0	47.0	17672	0.9433	0.8831
0.0	48.0	18048	0.9441	0.8815
0.0	49.0	18424	0.9445	0.8815
0.0047	50.0	18800	0.9436	0.8831

Framework versions

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