Simple implementation of finding the best split when building a tree for a random forest

random_forest.py

def find_better_split(self, var_idx):
  x,y = self.x.values[self.idxs,var_idx], self.y[self.idxs]
  
  for i in range(1,self.n-1):
    lhs = x<=x[i]
    rhs = x>x[i]
    if rhs.sum()==0: continue
    lhs_std = y[lhs].std()
    rhs_std = y[rhs].std()
    curr_score = lhs_std*lhs.sum() + rhs_std*rhs.sum()
    if curr_score<self.score: 
      self.var_idx,self.score,self.split = var_idx,curr_score,x[i]