smids_10x_deit_tiny_sgd_00001_fold2

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: 0.9487
• Accuracy: 0.5075

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
1.3101	1.0	750	1.2862	0.3394
1.2651	2.0	1500	1.2316	0.3494
1.2534	3.0	2250	1.1915	0.3461
1.2051	4.0	3000	1.1627	0.3627
1.1704	5.0	3750	1.1422	0.3644
1.1559	6.0	4500	1.1266	0.3727
1.1314	7.0	5250	1.1140	0.3827
1.0917	8.0	6000	1.1032	0.3844
1.0985	9.0	6750	1.0936	0.4193
1.0734	10.0	7500	1.0847	0.4243
1.0397	11.0	8250	1.0764	0.4293
1.0584	12.0	9000	1.0685	0.4409
1.0692	13.0	9750	1.0611	0.4426
1.0127	14.0	10500	1.0540	0.4542
1.0605	15.0	11250	1.0471	0.4542
1.0197	16.0	12000	1.0406	0.4626
1.0472	17.0	12750	1.0344	0.4659
0.9868	18.0	13500	1.0285	0.4709
1.0498	19.0	14250	1.0228	0.4725
0.9916	20.0	15000	1.0174	0.4742
1.0032	21.0	15750	1.0122	0.4792
1.0262	22.0	16500	1.0073	0.4792
0.9732	23.0	17250	1.0026	0.4792
0.9627	24.0	18000	0.9981	0.4875
0.9933	25.0	18750	0.9939	0.4892
0.9645	26.0	19500	0.9898	0.4942
0.9413	27.0	20250	0.9860	0.4942
0.9502	28.0	21000	0.9824	0.4925
0.9622	29.0	21750	0.9790	0.4958
0.9399	30.0	22500	0.9758	0.4975
0.9259	31.0	23250	0.9728	0.4992
0.9425	32.0	24000	0.9700	0.5008
0.9657	33.0	24750	0.9673	0.5042
0.9537	34.0	25500	0.9649	0.5058
0.9361	35.0	26250	0.9627	0.5075
0.934	36.0	27000	0.9606	0.5092
0.927	37.0	27750	0.9587	0.5092
0.9435	38.0	28500	0.9570	0.5092
0.9139	39.0	29250	0.9555	0.5092
0.9394	40.0	30000	0.9541	0.5075
0.9635	41.0	30750	0.9529	0.5075
0.9447	42.0	31500	0.9519	0.5075
0.9124	43.0	32250	0.9510	0.5075
0.9404	44.0	33000	0.9503	0.5075
0.9374	45.0	33750	0.9497	0.5075
0.9103	46.0	34500	0.9493	0.5075
0.9609	47.0	35250	0.9490	0.5075
0.9309	48.0	36000	0.9488	0.5075
0.9307	49.0	36750	0.9487	0.5075
0.9119	50.0	37500	0.9487	0.5075

Framework versions

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