smids_10x_deit_tiny_rms_00001_fold4

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: 1.3970
• Accuracy: 0.875

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
0.2502	1.0	750	0.3491	0.8717
0.1602	2.0	1500	0.3787	0.8617
0.0975	3.0	2250	0.4137	0.8717
0.0908	4.0	3000	0.4782	0.8767
0.061	5.0	3750	0.6214	0.88
0.0396	6.0	4500	0.7566	0.885
0.062	7.0	5250	0.8425	0.8683
0.0007	8.0	6000	0.8436	0.8817
0.0025	9.0	6750	0.9779	0.88
0.0255	10.0	7500	1.1344	0.8683
0.0352	11.0	8250	1.1423	0.8633
0.0318	12.0	9000	1.0833	0.865
0.0002	13.0	9750	1.1038	0.8767
0.0084	14.0	10500	1.0965	0.87
0.0351	15.0	11250	1.1367	0.8717
0.0252	16.0	12000	1.2686	0.875
0.0093	17.0	12750	1.0965	0.8833
0.0	18.0	13500	1.2326	0.8683
0.0001	19.0	14250	1.3027	0.8683
0.0	20.0	15000	1.1789	0.8783
0.0	21.0	15750	1.2125	0.875
0.0	22.0	16500	1.1748	0.8767
0.0001	23.0	17250	1.1790	0.87
0.0	24.0	18000	1.3742	0.8733
0.0	25.0	18750	1.2377	0.8783
0.0003	26.0	19500	1.3192	0.8783
0.0	27.0	20250	1.2703	0.8817
0.0001	28.0	21000	1.2600	0.8833
0.0	29.0	21750	1.2702	0.8867
0.0	30.0	22500	1.2442	0.8917
0.0	31.0	23250	1.2963	0.8817
0.0	32.0	24000	1.4012	0.88
0.0	33.0	24750	1.4514	0.8767
0.0	34.0	25500	1.4381	0.8733
0.0	35.0	26250	1.3220	0.8783
0.0	36.0	27000	1.3859	0.8767
0.0	37.0	27750	1.2544	0.8817
0.0	38.0	28500	1.2536	0.8817
0.0	39.0	29250	1.3812	0.88
0.0	40.0	30000	1.3350	0.8733
0.0	41.0	30750	1.4121	0.8733
0.0	42.0	31500	1.4110	0.8733
0.0	43.0	32250	1.4115	0.875
0.0	44.0	33000	1.3934	0.8783
0.0	45.0	33750	1.3917	0.88
0.0	46.0	34500	1.3899	0.88
0.0	47.0	35250	1.3925	0.88
0.0	48.0	36000	1.3926	0.88
0.0	49.0	36750	1.3955	0.875
0.0	50.0	37500	1.3970	0.875

Framework versions

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