smids_5x_deit_tiny_rms_001_fold3

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: 0.8753
• Accuracy: 0.7767

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
0.7906	1.0	375	0.9128	0.4983
0.7765	2.0	750	0.9232	0.4617
0.7977	3.0	1125	0.8743	0.5267
0.8093	4.0	1500	0.7926	0.5767
0.8508	5.0	1875	0.7894	0.5733
0.7532	6.0	2250	0.7991	0.6117
0.7584	7.0	2625	0.7566	0.625
0.7398	8.0	3000	0.7364	0.6083
0.7009	9.0	3375	0.7452	0.64
0.7014	10.0	3750	0.7192	0.6433
0.7226	11.0	4125	0.7119	0.6383
0.7293	12.0	4500	0.7180	0.6467
0.6344	13.0	4875	0.7612	0.6117
0.6251	14.0	5250	0.7810	0.66
0.6301	15.0	5625	0.6950	0.6733
0.6252	16.0	6000	0.7106	0.6767
0.688	17.0	6375	0.7082	0.6883
0.7261	18.0	6750	0.6859	0.6883
0.5633	19.0	7125	0.6734	0.7033
0.6092	20.0	7500	0.6580	0.7283
0.4728	21.0	7875	0.6793	0.7033
0.5681	22.0	8250	0.6598	0.7217
0.5951	23.0	8625	0.6134	0.7533
0.6592	24.0	9000	0.5954	0.7467
0.5215	25.0	9375	0.5847	0.74
0.5272	26.0	9750	0.6243	0.7017
0.5866	27.0	10125	0.6339	0.7233
0.5766	28.0	10500	0.5466	0.765
0.463	29.0	10875	0.5734	0.7583
0.5041	30.0	11250	0.5320	0.775
0.5133	31.0	11625	0.5507	0.7683
0.5402	32.0	12000	0.5711	0.7517
0.4526	33.0	12375	0.5736	0.7483
0.4724	34.0	12750	0.5009	0.79
0.3951	35.0	13125	0.5483	0.77
0.3876	36.0	13500	0.5689	0.755
0.3627	37.0	13875	0.5639	0.7733
0.4378	38.0	14250	0.5663	0.765
0.3725	39.0	14625	0.5574	0.7867
0.3444	40.0	15000	0.5740	0.7733
0.3158	41.0	15375	0.5671	0.7717
0.29	42.0	15750	0.6455	0.78
0.3784	43.0	16125	0.6093	0.785
0.318	44.0	16500	0.6835	0.7683
0.2949	45.0	16875	0.7092	0.7733
0.2996	46.0	17250	0.6699	0.7767
0.2938	47.0	17625	0.7545	0.7917
0.2248	48.0	18000	0.8050	0.775
0.2309	49.0	18375	0.8518	0.7767
0.1878	50.0	18750	0.8753	0.7767

Framework versions

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