Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/alexplatasl/dplbnde/llms.txt

Use this file to discover all available pages before exploring further.

All dplbnDE algorithm functions return an object of S3 class DE. Two S3 methods are provided for inspecting results.
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
FieldDescription
Number of evaluationsTotal CLL fitness calls made during evolution
Final population sizeNumber of individuals remaining (may be < NP for LPSR variants)
Best CLLHighest (least negative) Conditional Log-Likelihood found
Worst CLLLowest CLL in the final population
MedianMedian 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":
FieldTypeDescription
$Bestbnc_bnBest Bayesian Network found
$BestCLLnumericCLL of the best network
$pobFinallist of bnc_bnFinal population
$CLLPobFinalnumeric vectorCLL of each final individual
$N.evalsintegerTotal evaluations
$convergencenumeric vectorBest CLL per generation
$evaluationsinteger vectorCumulative evaluations per generation
Use $Best with bnclassify for prediction:
predictions <- predict(result$Best, newdata)

Build docs developers (and LLMs) love