smids_3x_deit_tiny_rms_00001_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: 0.9365
• 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
0.3997	1.0	226	0.3347	0.8631
0.255	2.0	452	0.2918	0.8831
0.1424	3.0	678	0.3193	0.8748
0.1606	4.0	904	0.3236	0.8865
0.1532	5.0	1130	0.3062	0.8915
0.0588	6.0	1356	0.4276	0.8915
0.0136	7.0	1582	0.4629	0.8831
0.0352	8.0	1808	0.5602	0.8765
0.0656	9.0	2034	0.5379	0.8765
0.0146	10.0	2260	0.6661	0.8881
0.0002	11.0	2486	0.7507	0.8798
0.015	12.0	2712	0.6981	0.8865
0.0138	13.0	2938	0.9249	0.8715
0.0124	14.0	3164	0.8454	0.8748
0.0002	15.0	3390	0.8233	0.8781
0.0006	16.0	3616	0.8574	0.8698
0.0171	17.0	3842	0.8765	0.8781
0.0	18.0	4068	0.8826	0.8865
0.0173	19.0	4294	0.7556	0.8932
0.0158	20.0	4520	0.9424	0.8748
0.0001	21.0	4746	1.0298	0.8648
0.0133	22.0	4972	0.9420	0.8664
0.0145	23.0	5198	0.8626	0.8865
0.0001	24.0	5424	0.9250	0.8781
0.0	25.0	5650	0.8112	0.8948
0.0002	26.0	5876	0.8569	0.8898
0.0	27.0	6102	0.8070	0.8915
0.0	28.0	6328	0.8507	0.8765
0.0	29.0	6554	0.8437	0.8932
0.0	30.0	6780	0.8816	0.8848
0.0	31.0	7006	0.8733	0.8848
0.0	32.0	7232	0.9948	0.8681
0.0082	33.0	7458	0.9148	0.8831
0.0	34.0	7684	0.9131	0.8881
0.0	35.0	7910	0.9403	0.8781
0.0	36.0	8136	0.9014	0.8765
0.0	37.0	8362	0.9056	0.8798
0.0	38.0	8588	0.9375	0.8781
0.0	39.0	8814	0.9025	0.8831
0.0	40.0	9040	0.9205	0.8815
0.0051	41.0	9266	0.9089	0.8848
0.0025	42.0	9492	0.9223	0.8848
0.0	43.0	9718	0.9136	0.8881
0.0	44.0	9944	0.9207	0.8848
0.0	45.0	10170	0.9266	0.8831
0.0	46.0	10396	0.9325	0.8865
0.0	47.0	10622	0.9382	0.8815
0.0	48.0	10848	0.9372	0.8815
0.0	49.0	11074	0.9372	0.8831
0.0	50.0	11300	0.9365	0.8831

Framework versions

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