search

timmens / causal-forest

Implements the Causal Forest algorithm formulated in Athey and Wager (2018).
MIT License
65 stars 12 forks source link

Add random feature selection to tree building in forest. #13

Closed timmens closed 4 years ago

timmens commented 4 years ago

At the moment we consider all features for the tree building process. Goal:

  1. Use parameter ratio_features_at_split to determine the ratio of features to consider at each splitting point in a tree. -> rewrite _fit method in causaltree.py.
  2. Check if number of features is below a threshold (< 3) then do not use the ratio but use all variables and print a warning message.
timmens commented 4 years ago

Issue was solved/added by a7939d7fe07a0a.