Open HeuristicLab-Trac-Bot opened 5 years ago
Issue migrated from trac ticket # 2990
milestone: HeuristicLab 3.3.17 | component: Algorithms.DataAnalysis | priority: medium
Using the variable impacts for iterative feature selection can currently only be done manually. Having an automatic way would be nice. General workflow of this Variable-Impact-based Feature Selection: Train Model Calculate Variable Impacts Remove Features with low impact Repeat until model accuracy drops too much
Using the variable impacts for iterative feature selection can currently only be done manually. Having an automatic way would be nice.
General workflow of this Variable-Impact-based Feature Selection:
r16612 branched trunk
r16705: Implemented first version with RF and percentage-based feature elimination.
Issue migrated from trac ticket # 2990
milestone: HeuristicLab 3.3.17 | component: Algorithms.DataAnalysis | priority: medium
2019-02-17 12:02:43: @NimZwei created the issue