Latest version: v0.0.15
The information on this page was curated by experts in our Cybersecurity Intelligence Team.
Combines feature selection, model tuning, and parsimonious model selection with GA optimization. GA selection procedure is based on separate cost and complexity evaluations. Therefore, the best individuals are initially sorted by an error fitness function, and afterwards, models with similar costs are rearranged according to modelcomplexity measurement so as to foster models of lesser complexity. The algorithm can be run sequentially or in parallel.
No known vulnerabilities found
Has known vulnerabilities