smids_10x_deit_tiny_rms_00001_fold5

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: 1.0880
• Accuracy: 0.9017

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
0.2267	1.0	750	0.2310	0.9017
0.1688	2.0	1500	0.3133	0.8767
0.1207	3.0	2250	0.3307	0.8917
0.0544	4.0	3000	0.3393	0.8933
0.0685	5.0	3750	0.4794	0.8817
0.0086	6.0	4500	0.5790	0.8967
0.0247	7.0	5250	0.6674	0.895
0.0176	8.0	6000	0.8292	0.89
0.0446	9.0	6750	0.8047	0.8933
0.0267	10.0	7500	0.8263	0.8983
0.0376	11.0	8250	0.8873	0.9
0.0007	12.0	9000	0.9745	0.8883
0.0033	13.0	9750	0.8924	0.9033
0.0025	14.0	10500	0.8544	0.905
0.0004	15.0	11250	1.0444	0.8967
0.0002	16.0	12000	1.0224	0.8933
0.0105	17.0	12750	1.0131	0.8883
0.014	18.0	13500	0.9525	0.9033
0.0	19.0	14250	0.9850	0.8983
0.0	20.0	15000	1.1461	0.89
0.003	21.0	15750	0.9659	0.9033
0.0004	22.0	16500	1.0728	0.8967
0.0257	23.0	17250	1.0751	0.9
0.006	24.0	18000	1.1593	0.8933
0.0242	25.0	18750	1.1220	0.8917
0.0	26.0	19500	1.0931	0.9033
0.0	27.0	20250	1.0693	0.9017
0.0	28.0	21000	1.2173	0.88
0.0	29.0	21750	1.0569	0.905
0.0118	30.0	22500	1.1864	0.8917
0.0	31.0	23250	1.2141	0.8967
0.0	32.0	24000	1.1888	0.8933
0.0	33.0	24750	1.1513	0.8983
0.0	34.0	25500	1.2676	0.89
0.0	35.0	26250	1.1568	0.8983
0.0	36.0	27000	1.1800	0.89
0.0068	37.0	27750	1.1482	0.9033
0.0	38.0	28500	1.0989	0.9
0.0	39.0	29250	1.1226	0.9
0.0	40.0	30000	1.1146	0.8983
0.0	41.0	30750	1.0950	0.9
0.0	42.0	31500	1.0906	0.9
0.0	43.0	32250	1.0973	0.9
0.0	44.0	33000	1.0952	0.9
0.0	45.0	33750	1.0896	0.8983
0.0	46.0	34500	1.0935	0.8983
0.0	47.0	35250	1.0921	0.8983
0.0	48.0	36000	1.0899	0.8983
0.0	49.0	36750	1.0869	0.9017
0.0	50.0	37500	1.0880	0.9017

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2