Table 2.

Timings (seconds) for logistic models with lasso penalty. Total time for tenfold cross-validation over a grid of 100 λ values.

	Logistic Regression — Dense Features
	Correlation
	0	0.1	0.2	0.5	0.9	0.95
	N = 1000, p = 100
glmnet	1.65	1.81	2.31	3.87	5.99	8.48
l1lognet	31.475	31.86	34.35	32.21	31.85	31.81
BBR	40.70	47.57	54.18	70.06	106.72	121.41
LPL	24.68	31.64	47.99	170.77	741.00	1448.25
	N = 5000, p = 100
glmnet	7.89	8.48	9.01	13.39	26.68	26.36
l1lognet	239.88	232.00	229.62	229.49	22.19	223.09
	N = 100, 000, p = 100
glmnet	78.56	178.45	205.94	274.33	552.48	638.50
	N = 100, p = 1000
glmnet	1.06	1.07	1.09	1.45	1.72	1.37
l1lognet	25.99	26.40	25.67	26.49	24.34	20.16
BBR	70.19	71.19	78.40	103.77	149.05	113.87
LPL	11.02	10.87	10.76	16.34	41.84	70.50
	N = 100, p = 5000
glmnet	5.24	4.43	5.12	7.05	7.87	6.05
l1lognet	165.02	161.90	163.25	166.50	151.91	135.28
	N = 100, p = 100, 000
glmnet	137.27	139.40	146.55	197.98	219.65	201.93