smids_10x_deit_small_rms_0001_fold5

This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.1181
• Accuracy: 0.8983

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
0.1916	1.0	750	0.2932	0.8983
0.1713	2.0	1500	0.3029	0.9083
0.1109	3.0	2250	0.5067	0.8767
0.0732	4.0	3000	0.4780	0.91
0.102	5.0	3750	0.4476	0.8983
0.0951	6.0	4500	0.5447	0.9017
0.0348	7.0	5250	0.5626	0.915
0.0144	8.0	6000	0.5893	0.9133
0.0592	9.0	6750	0.5568	0.91
0.0154	10.0	7500	0.5698	0.9017
0.0271	11.0	8250	0.6256	0.8983
0.0176	12.0	9000	0.7391	0.8917
0.012	13.0	9750	0.6256	0.8967
0.0094	14.0	10500	0.7473	0.895
0.0301	15.0	11250	0.6066	0.905
0.0069	16.0	12000	0.6970	0.9
0.0131	17.0	12750	0.6902	0.895
0.0022	18.0	13500	0.7962	0.8833
0.0162	19.0	14250	0.8033	0.8967
0.0045	20.0	15000	0.7612	0.8933
0.0063	21.0	15750	0.7939	0.9
0.0002	22.0	16500	0.7612	0.8933
0.0161	23.0	17250	0.8161	0.8867
0.0001	24.0	18000	0.8196	0.8933
0.038	25.0	18750	0.8702	0.8883
0.0257	26.0	19500	0.7862	0.8983
0.0004	27.0	20250	0.8138	0.8983
0.0001	28.0	21000	0.8830	0.905
0.0003	29.0	21750	1.0169	0.89
0.016	30.0	22500	0.8531	0.8883
0.0027	31.0	23250	0.9699	0.895
0.0379	32.0	24000	1.0313	0.89
0.0307	33.0	24750	0.8698	0.905
0.0	34.0	25500	0.8949	0.9017
0.0	35.0	26250	0.9260	0.8917
0.0	36.0	27000	0.9677	0.89
0.0036	37.0	27750	1.0175	0.9017
0.0	38.0	28500	1.0579	0.8967
0.0003	39.0	29250	1.1044	0.8917
0.0	40.0	30000	1.0357	0.895
0.0	41.0	30750	1.0321	0.9
0.0	42.0	31500	1.0802	0.895
0.0	43.0	32250	1.1064	0.895
0.0	44.0	33000	1.1154	0.9017
0.0	45.0	33750	1.1069	0.9017
0.0	46.0	34500	1.1200	0.895
0.0	47.0	35250	1.1205	0.8967
0.0	48.0	36000	1.1169	0.8983
0.0	49.0	36750	1.1164	0.8983
0.0	50.0	37500	1.1181	0.8983

Framework versions

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