smids_3x_deit_small_sgd_0001_fold1

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: 0.5893
• Accuracy: 0.7546

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
1.0679	1.0	226	1.0513	0.4491
1.0307	2.0	452	1.0262	0.4841
0.9834	3.0	678	1.0025	0.5309
0.9773	4.0	904	0.9789	0.5526
0.9692	5.0	1130	0.9560	0.5710
0.9555	6.0	1356	0.9334	0.5843
0.9069	7.0	1582	0.9110	0.6077
0.8993	8.0	1808	0.8899	0.6210
0.8781	9.0	2034	0.8696	0.6327
0.881	10.0	2260	0.8502	0.6511
0.8393	11.0	2486	0.8314	0.6661
0.8206	12.0	2712	0.8136	0.6694
0.7291	13.0	2938	0.7969	0.6661
0.7685	14.0	3164	0.7808	0.6728
0.7796	15.0	3390	0.7656	0.6795
0.7626	16.0	3616	0.7511	0.6845
0.6907	17.0	3842	0.7376	0.6962
0.7127	18.0	4068	0.7250	0.7028
0.7138	19.0	4294	0.7133	0.7078
0.6728	20.0	4520	0.7022	0.7179
0.6631	21.0	4746	0.6922	0.7229
0.6986	22.0	4972	0.6830	0.7312
0.6948	23.0	5198	0.6744	0.7346
0.6653	24.0	5424	0.6664	0.7396
0.6494	25.0	5650	0.6590	0.7446
0.68	26.0	5876	0.6520	0.7462
0.5871	27.0	6102	0.6459	0.7462
0.6576	28.0	6328	0.6401	0.7479
0.5996	29.0	6554	0.6347	0.7496
0.6519	30.0	6780	0.6298	0.7496
0.6001	31.0	7006	0.6252	0.7496
0.6281	32.0	7232	0.6210	0.7479
0.555	33.0	7458	0.6170	0.7479
0.5831	34.0	7684	0.6134	0.7513
0.6219	35.0	7910	0.6102	0.7496
0.5515	36.0	8136	0.6072	0.7496
0.5829	37.0	8362	0.6045	0.7513
0.6171	38.0	8588	0.6020	0.7513
0.5762	39.0	8814	0.5998	0.7513
0.5768	40.0	9040	0.5978	0.7513
0.5572	41.0	9266	0.5960	0.7513
0.5838	42.0	9492	0.5945	0.7513
0.5735	43.0	9718	0.5931	0.7513
0.5285	44.0	9944	0.5920	0.7513
0.6211	45.0	10170	0.5911	0.7513
0.5665	46.0	10396	0.5904	0.7546
0.5672	47.0	10622	0.5899	0.7563
0.562	48.0	10848	0.5895	0.7546
0.5485	49.0	11074	0.5894	0.7546
0.5589	50.0	11300	0.5893	0.7546

Framework versions

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