smids_10x_deit_tiny_rms_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: 1.2287
• Accuracy: 0.8869

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
0.2512	1.0	750	0.3089	0.8735
0.118	2.0	1500	0.3336	0.8819
0.0986	3.0	2250	0.3751	0.8735
0.1049	4.0	3000	0.4900	0.8819
0.0487	5.0	3750	0.5969	0.8835
0.0234	6.0	4500	0.6491	0.8835
0.036	7.0	5250	0.7457	0.8869
0.001	8.0	6000	0.8757	0.8719
0.0148	9.0	6750	0.8581	0.8869
0.0298	10.0	7500	1.0727	0.8719
0.0019	11.0	8250	1.0234	0.8719
0.0146	12.0	9000	1.1199	0.8702
0.0069	13.0	9750	1.0417	0.8785
0.0001	14.0	10500	1.0745	0.8819
0.0	15.0	11250	1.0294	0.8802
0.0091	16.0	12000	1.1025	0.8802
0.0186	17.0	12750	1.0736	0.8835
0.0	18.0	13500	1.0297	0.8769
0.047	19.0	14250	1.1126	0.8819
0.0	20.0	15000	1.1842	0.8785
0.0	21.0	15750	1.2771	0.8686
0.0198	22.0	16500	1.1311	0.8869
0.0357	23.0	17250	1.1425	0.8869
0.0	24.0	18000	1.1413	0.8885
0.0	25.0	18750	1.1558	0.8852
0.0	26.0	19500	1.1246	0.8869
0.0271	27.0	20250	1.2507	0.8752
0.0	28.0	21000	1.1107	0.8902
0.0	29.0	21750	1.1979	0.8852
0.0	30.0	22500	1.2404	0.8869
0.0	31.0	23250	1.2332	0.8819
0.0	32.0	24000	1.3008	0.8819
0.0	33.0	24750	1.3101	0.8819
0.0	34.0	25500	1.3030	0.8869
0.0	35.0	26250	1.1931	0.8835
0.0	36.0	27000	1.2127	0.8802
0.0203	37.0	27750	1.1903	0.8802
0.0	38.0	28500	1.2617	0.8869
0.0	39.0	29250	1.1890	0.8935
0.0	40.0	30000	1.2276	0.8869
0.0	41.0	30750	1.1665	0.8885
0.0	42.0	31500	1.1610	0.8852
0.0	43.0	32250	1.2105	0.8902
0.0	44.0	33000	1.2243	0.8885
0.0031	45.0	33750	1.2267	0.8902
0.0	46.0	34500	1.2227	0.8885
0.0	47.0	35250	1.2254	0.8869
0.0	48.0	36000	1.2244	0.8852
0.0	49.0	36750	1.2292	0.8869
0.0	50.0	37500	1.2287	0.8869

Framework versions

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