smids_3x_deit_tiny_rms_0001_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: 0.8923
• Accuracy: 0.8982

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: 0.0001
• 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
0.4584	1.0	226	0.3564	0.8414
0.3047	2.0	452	0.3557	0.8614
0.2416	3.0	678	0.4578	0.8097
0.1428	4.0	904	0.4233	0.8631
0.1227	5.0	1130	0.4440	0.8698
0.1524	6.0	1356	0.3981	0.8915
0.0691	7.0	1582	0.5148	0.8932
0.0615	8.0	1808	0.6305	0.8581
0.1111	9.0	2034	0.5927	0.8848
0.0229	10.0	2260	0.6397	0.8932
0.0429	11.0	2486	0.8816	0.8497
0.0206	12.0	2712	0.6220	0.8815
0.0864	13.0	2938	0.9519	0.8481
0.0084	14.0	3164	0.6931	0.8898
0.0377	15.0	3390	0.7392	0.8815
0.0362	16.0	3616	0.8551	0.8581
0.0572	17.0	3842	0.7247	0.8698
0.1009	18.0	4068	0.6228	0.8715
0.0504	19.0	4294	0.8902	0.8648
0.0251	20.0	4520	0.7786	0.8831
0.0065	21.0	4746	0.9795	0.8781
0.0845	22.0	4972	0.7973	0.8815
0.0036	23.0	5198	0.8709	0.8648
0.0477	24.0	5424	0.8349	0.8781
0.0002	25.0	5650	0.8199	0.8781
0.0014	26.0	5876	1.0541	0.8648
0.0123	27.0	6102	0.9016	0.8798
0.0001	28.0	6328	0.7666	0.8881
0.0	29.0	6554	0.7915	0.8881
0.0004	30.0	6780	0.9036	0.8881
0.0001	31.0	7006	0.8349	0.8932
0.0	32.0	7232	0.8309	0.8848
0.0025	33.0	7458	0.8690	0.8932
0.0	34.0	7684	0.9464	0.8848
0.0312	35.0	7910	0.9721	0.8881
0.0	36.0	8136	0.9282	0.8848
0.0	37.0	8362	0.8805	0.8915
0.0	38.0	8588	0.8709	0.8948
0.0	39.0	8814	0.8888	0.8915
0.0	40.0	9040	0.8303	0.9015
0.0028	41.0	9266	0.8497	0.8965
0.0031	42.0	9492	0.8601	0.8932
0.0	43.0	9718	0.8400	0.8932
0.0	44.0	9944	0.8634	0.8965
0.0	45.0	10170	0.8516	0.9015
0.0	46.0	10396	0.9026	0.8948
0.0	47.0	10622	0.8950	0.9015
0.0	48.0	10848	0.8925	0.8998
0.0	49.0	11074	0.8907	0.8982
0.0	50.0	11300	0.8923	0.8982

Framework versions

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