Journal of Inequalities and Applications

Table 9 Comparison of each algorithm at the 100th iteration with 10-fold CV. on Iris dataset

From: On convergence and complexity analysis of an accelerated forward–backward algorithm with linesearch technique for convex minimization problems and applications to data prediction and classification

	Algorithm 1		Algorithm 2		Algorithm 3
	Acc. train	Acc. test	Acc. train	Acc. test	Acc. train	Acc. test
Fold 1	83.703	86.666	84.444	86.666	97.04	93.33
Fold 2	84.444	80	85.925	80	97.78	86.67
Fold 3	82.962	93.333	84.444	93.333	94.81	93.33
Fold 4	82.962	93.333	84.444	93.333	97.04	93.33
Fold 5	84.444	80	85.185	80	97.04	100
Fold 6	85.185	73.333	85.925	73.333	94.81	93.33
Fold 7	82.222	93.333	83.703	93.333	96.30	86.67
Fold 8	82.962	86.666	84.444	86.666	97.04	100
Fold 9	84.444	80	85.185	80	94.81	100.00
Fold 10	85.185	73.333	85.925	73.333	96.30	100
Average Acc	83.851	84	84.962	84	96.30	94.67
\(\operatorname{ERR} _{ \% }\)	16.074		15.5185		4.52
Training time (sec.)	0.0199		0.0329		0.0276

Back to article page