convnext-tiny-224-driverbox

This model is a fine-tuned version of facebook/convnext-tiny-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.3349	0.9950	99	0.2700	0.9328
0.2393	2.0	199	0.1932	0.9540
0.1831	2.9950	298	0.1403	0.9618
0.1397	4.0	398	0.1055	0.9689
0.0795	4.9950	497	0.1030	0.9731
0.0915	6.0	597	0.0966	0.9703
0.0718	6.9950	696	0.0779	0.9745
0.0502	8.0	796	0.0729	0.9788
0.0314	8.9950	895	0.0621	0.9802
0.0408	10.0	995	0.0758	0.9752
0.0335	10.9950	1094	0.0598	0.9823
0.0228	12.0	1194	0.0573	0.9823
0.0229	12.9950	1293	0.0473	0.9844
0.0119	14.0	1393	0.0642	0.9844
0.028	14.9950	1492	0.0526	0.9851
0.0117	16.0	1592	0.0594	0.9837
0.0187	16.9950	1691	0.0497	0.9880
0.0131	18.0	1791	0.0663	0.9837
0.0132	18.9950	1890	0.0478	0.9866
0.014	20.0	1990	0.0465	0.9880
0.0039	20.9950	2089	0.0496	0.9851
0.0102	22.0	2189	0.0468	0.9880
0.0035	22.9950	2288	0.0581	0.9866
0.0071	24.0	2388	0.0519	0.9866
0.0032	24.9950	2487	0.0510	0.9880
0.0049	26.0	2587	0.0575	0.9858
0.0037	26.9950	2686	0.0511	0.9880
0.0029	28.0	2786	0.0484	0.9880
0.0019	28.9950	2885	0.0523	0.9866
0.0058	29.8492	2970	0.0532	0.9866