smids_10x_deit_small_rms_001_fold3

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: 0.5590
• 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
0.8839	1.0	750	0.8956	0.4917
0.8402	2.0	1500	0.8459	0.5383
0.827	3.0	2250	0.8365	0.5417
0.7595	4.0	3000	0.8404	0.5617
0.8496	5.0	3750	0.9112	0.505
0.7825	6.0	4500	0.8246	0.6233
0.8185	7.0	5250	0.7843	0.6233
0.7863	8.0	6000	0.7862	0.6183
0.7304	9.0	6750	0.7478	0.6433
0.7486	10.0	7500	0.7941	0.625
0.7979	11.0	8250	0.7438	0.6817
0.6928	12.0	9000	0.8898	0.58
0.683	13.0	9750	0.7126	0.68
0.7194	14.0	10500	0.7634	0.6367
0.7001	15.0	11250	0.6906	0.68
0.7209	16.0	12000	0.6988	0.675
0.693	17.0	12750	0.7227	0.6733
0.6594	18.0	13500	0.7119	0.675
0.6733	19.0	14250	0.6769	0.695
0.6368	20.0	15000	0.6310	0.7183
0.5529	21.0	15750	0.6379	0.73
0.674	22.0	16500	0.6200	0.7233
0.6173	23.0	17250	0.6390	0.7117
0.7017	24.0	18000	0.6234	0.7217
0.6672	25.0	18750	0.6159	0.7117
0.6143	26.0	19500	0.6119	0.7133
0.5447	27.0	20250	0.6511	0.7
0.616	28.0	21000	0.5943	0.7317
0.6257	29.0	21750	0.6135	0.7417
0.5784	30.0	22500	0.6236	0.7383
0.5488	31.0	23250	0.5814	0.7483
0.5683	32.0	24000	0.6409	0.725
0.5657	33.0	24750	0.6193	0.7583
0.7061	34.0	25500	0.7958	0.6533
0.5815	35.0	26250	0.6092	0.7467
0.545	36.0	27000	0.5902	0.7567
0.574	37.0	27750	0.5865	0.7483
0.5654	38.0	28500	0.6161	0.7467
0.5393	39.0	29250	0.5677	0.7667
0.6213	40.0	30000	0.5702	0.7633
0.5565	41.0	30750	0.5675	0.75
0.5323	42.0	31500	0.5645	0.7583
0.5444	43.0	32250	0.5820	0.76
0.4988	44.0	33000	0.5588	0.765
0.5249	45.0	33750	0.5669	0.7583
0.5246	46.0	34500	0.5504	0.7733
0.4975	47.0	35250	0.5697	0.7717
0.5083	48.0	36000	0.5554	0.7717
0.4948	49.0	36750	0.5551	0.775
0.4147	50.0	37500	0.5590	0.7767

Framework versions

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