smids_3x_deit_tiny_adamax_00001_fold1

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: 0.9289
• Accuracy: 0.8715

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
0.49	1.0	226	0.4492	0.8013
0.3613	2.0	452	0.3379	0.8614
0.2434	3.0	678	0.3074	0.8648
0.2553	4.0	904	0.3243	0.8648
0.2473	5.0	1130	0.2827	0.8831
0.1686	6.0	1356	0.3078	0.8765
0.1222	7.0	1582	0.3023	0.8998
0.1406	8.0	1808	0.3325	0.8865
0.0989	9.0	2034	0.3862	0.8798
0.0281	10.0	2260	0.3985	0.8748
0.0373	11.0	2486	0.4395	0.8831
0.0324	12.0	2712	0.4479	0.8898
0.0085	13.0	2938	0.5150	0.8865
0.0336	14.0	3164	0.5239	0.8831
0.0184	15.0	3390	0.5580	0.8798
0.0187	16.0	3616	0.6394	0.8798
0.0341	17.0	3842	0.7055	0.8715
0.0009	18.0	4068	0.6833	0.8698
0.0242	19.0	4294	0.6897	0.8731
0.0002	20.0	4520	0.7463	0.8715
0.0021	21.0	4746	0.7865	0.8664
0.0168	22.0	4972	0.7905	0.8715
0.0077	23.0	5198	0.7986	0.8715
0.0002	24.0	5424	0.8358	0.8715
0.0002	25.0	5650	0.8300	0.8698
0.0001	26.0	5876	0.8435	0.8681
0.0001	27.0	6102	0.8418	0.8681
0.0001	28.0	6328	0.8696	0.8681
0.0	29.0	6554	0.8706	0.8698
0.0001	30.0	6780	0.9033	0.8698
0.0001	31.0	7006	0.9296	0.8681
0.0001	32.0	7232	0.8999	0.8698
0.0096	33.0	7458	0.9062	0.8681
0.0001	34.0	7684	0.9009	0.8715
0.0	35.0	7910	0.8975	0.8765
0.0	36.0	8136	0.9003	0.8748
0.0	37.0	8362	0.9103	0.8731
0.0	38.0	8588	0.9226	0.8664
0.0	39.0	8814	0.9185	0.8698
0.0	40.0	9040	0.9208	0.8715
0.0079	41.0	9266	0.9347	0.8698
0.0103	42.0	9492	0.9073	0.8731
0.0	43.0	9718	0.9457	0.8664
0.0	44.0	9944	0.9277	0.8698
0.0	45.0	10170	0.9217	0.8715
0.0	46.0	10396	0.9203	0.8715
0.0	47.0	10622	0.9223	0.8715
0.0	48.0	10848	0.9286	0.8715
0.0	49.0	11074	0.9289	0.8715
0.0	50.0	11300	0.9289	0.8715

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.1+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2