smids_5x_deit_tiny_rms_0001_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: 1.1259
• Accuracy: 0.8735

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
0.3139	1.0	375	0.2920	0.8835
0.213	2.0	750	0.3450	0.8785
0.2004	3.0	1125	0.4306	0.8719
0.1151	4.0	1500	0.4856	0.8702
0.1363	5.0	1875	0.5483	0.8752
0.0415	6.0	2250	0.6014	0.8719
0.0888	7.0	2625	0.6594	0.8636
0.0129	8.0	3000	0.7394	0.8702
0.0606	9.0	3375	0.7551	0.8619
0.0273	10.0	3750	0.7977	0.8536
0.0575	11.0	4125	0.7927	0.8702
0.0142	12.0	4500	0.8285	0.8619
0.006	13.0	4875	0.8594	0.8819
0.0339	14.0	5250	0.8600	0.8686
0.0029	15.0	5625	0.9289	0.8719
0.0348	16.0	6000	0.7828	0.8819
0.0273	17.0	6375	0.7381	0.8885
0.029	18.0	6750	0.9087	0.8686
0.0306	19.0	7125	0.9194	0.8785
0.0034	20.0	7500	1.0978	0.8619
0.0052	21.0	7875	0.9530	0.8785
0.0001	22.0	8250	0.9575	0.8752
0.0447	23.0	8625	0.9869	0.8819
0.0122	24.0	9000	0.8869	0.8785
0.0018	25.0	9375	1.0324	0.8669
0.0117	26.0	9750	0.9387	0.8852
0.0206	27.0	10125	1.0468	0.8719
0.0002	28.0	10500	0.9421	0.8785
0.0001	29.0	10875	0.8621	0.8968
0.0027	30.0	11250	0.9653	0.8769
0.0116	31.0	11625	0.9958	0.8785
0.0019	32.0	12000	1.1300	0.8752
0.0084	33.0	12375	1.0346	0.8802
0.0	34.0	12750	1.0458	0.8719
0.0	35.0	13125	1.0740	0.8719
0.0001	36.0	13500	1.0706	0.8719
0.0	37.0	13875	1.2116	0.8735
0.0	38.0	14250	1.1598	0.8735
0.0	39.0	14625	1.1682	0.8785
0.0029	40.0	15000	1.0573	0.8835
0.0	41.0	15375	1.1307	0.8735
0.0028	42.0	15750	1.1484	0.8702
0.0032	43.0	16125	1.1289	0.8752
0.0031	44.0	16500	1.1224	0.8769
0.0027	45.0	16875	1.1287	0.8719
0.0	46.0	17250	1.1176	0.8752
0.006	47.0	17625	1.1207	0.8752
0.0	48.0	18000	1.1234	0.8752
0.0024	49.0	18375	1.1256	0.8752
0.0022	50.0	18750	1.1259	0.8735

Framework versions

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