Skip to main content

Table 1 Performance evaluation

From: Seagull: lasso, group lasso and sparse-group lasso regularization for linear regression models via proximal gradient descent

R package Accuracy parameter R2 MSE Non-zero Time
SGL 10−4 0.91 12.78 8822 45 h 20 min
seagull 10−4 0.92 12.38 65,463 20 min
seagull 10−5 0.92 11.57 11,823 40 min
seagull 10−6 0.92 11.79 5095 2 h 13 min
seagull 10−8 0.92 11.84 5072 4 h 50 min
  1. Accuracy parameter refers to a package-dependent convergence parameter; R2 is the squared correlation coefficient and MSE is the mean squared error of chronological and predicted age; Non-zero denotes the number of CpG sites with non-zero effect estimate; Time is the computational time needed to calculate the full regularization path