DE class — print and plot

S3 methods for the DE result object: print.DE and plot.DE

All dplbnDE algorithm functions return an object of S3 class DE. Two S3 methods are provided for inspecting results.

print.DE

print(x, ...)

Prints a summary of the evolution results.

Output

Number of evaluations: 	 780
Final population size: 	 30

Summary results of fitness in final population:

Best CLL: 	 -1181.117
Worst CLL: 	 -1251.721
Median: 	 -1218.063
Std. Dev.: 	 17.96752

Field	Description
Number of evaluations	Total CLL fitness calls made during evolution
Final population size	Number of individuals remaining (may be < NP for LPSR variants)
Best CLL	Highest (least negative) Conditional Log-Likelihood found
Worst CLL	Lowest CLL in the final population
Median	Median CLL across final population
Std. Dev.	Standard deviation of CLL — low values indicate convergence

plot.DE

plot(x, ...)

Produces two side-by-side panels:

1. CLL histogram — distribution of CLL values across the final population. A narrow, peaked histogram indicates the population has converged to a high-quality region.
2. Convergence plot — best CLL vs. cumulative evaluations. A flattening curve indicates convergence; a still-declining curve suggests increasing G would help.

Example

library(dplbnDE)
data(car)

result <- lshade(NP=20, G=25, data=car, class.name=names(car)[7],
                 structure="tan", verbose=0)

print(result)
plot(result)

DE object fields

All algorithm functions return a list with class "DE":

Field	Type	Description
$Best	bnc_bn	Best Bayesian Network found
$BestCLL	numeric	CLL of the best network
$pobFinal	list of bnc_bn	Final population
$CLLPobFinal	numeric vector	CLL of each final individual
$N.evals	integer	Total evaluations
$convergence	numeric vector	Best CLL per generation
$evaluations	integer vector	Cumulative evaluations per generation

Use $Best with bnclassify for prediction:

predictions <- predict(result$Best, newdata)