smids_10x_deit_tiny_adamax_001_fold1

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

• Loss: 1.1377
• Accuracy: 0.8948

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
0.387	1.0	751	0.3856	0.8447
0.2275	2.0	1502	0.3799	0.8497
0.1732	3.0	2253	0.3628	0.8898
0.1418	4.0	3004	0.3720	0.8848
0.1851	5.0	3755	0.4163	0.8497
0.1168	6.0	4506	0.4228	0.8915
0.1217	7.0	5257	0.4050	0.8965
0.0972	8.0	6008	0.4659	0.8881
0.0717	9.0	6759	0.4692	0.8848
0.0615	10.0	7510	0.5939	0.8748
0.0582	11.0	8261	0.5202	0.8898
0.0569	12.0	9012	0.5681	0.8982
0.0142	13.0	9763	0.7223	0.8815
0.0849	14.0	10514	0.6292	0.8948
0.0289	15.0	11265	0.7113	0.8898
0.0438	16.0	12016	0.6702	0.8982
0.0561	17.0	12767	0.7629	0.8765
0.0013	18.0	13518	0.7639	0.8865
0.0173	19.0	14269	0.6756	0.8965
0.0044	20.0	15020	0.7365	0.8965
0.013	21.0	15771	0.8044	0.8831
0.0056	22.0	16522	0.7938	0.8915
0.0006	23.0	17273	0.8954	0.8848
0.0157	24.0	18024	0.8083	0.8998
0.0002	25.0	18775	0.8156	0.8965
0.0001	26.0	19526	0.8204	0.8982
0.0087	27.0	20277	0.8556	0.8948
0.0001	28.0	21028	0.8189	0.9048
0.0132	29.0	21779	0.8401	0.9065
0.0001	30.0	22530	0.9274	0.8915
0.0	31.0	23281	0.9668	0.8965
0.0153	32.0	24032	0.9746	0.8932
0.0	33.0	24783	1.0269	0.8881
0.0	34.0	25534	1.0125	0.8948
0.0	35.0	26285	1.0419	0.8898
0.0003	36.0	27036	1.0764	0.8898
0.0	37.0	27787	1.0824	0.8915
0.0	38.0	28538	1.0882	0.8898
0.0	39.0	29289	1.0563	0.8932
0.0	40.0	30040	1.0771	0.8915
0.0	41.0	30791	1.0705	0.8948
0.0	42.0	31542	1.0752	0.8932
0.0	43.0	32293	1.1011	0.8948
0.0	44.0	33044	1.1049	0.8948
0.0	45.0	33795	1.1132	0.8948
0.0	46.0	34546	1.1208	0.8965
0.0	47.0	35297	1.1280	0.8948
0.0	48.0	36048	1.1328	0.8948
0.0	49.0	36799	1.1361	0.8948
0.0	50.0	37550	1.1377	0.8948

Framework versions

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