:
这里有一段用决策树对变量进行分箱的python代码:
def binning(x, y, n_bins): # 将特征x和标签y分成n_bins组 bins = np.linspace(x.min(), x.max(), n_bins+1) bin_idx = np.digitize(x, bins) bin_means_y = [y[bin_idx == i].mean() for i in range(1, n_bins+1)] return bins, bin_means_y
:
这里有一段用决策树对变量进行分箱的python代码:
def binning(x, y, n_bins): # 将特征x和标签y分成n_bins组 bins = np.linspace(x.min(), x.max(), n_bins+1) bin_idx = np.digitize(x, bins) bin_means_y = [y[bin_idx == i].mean() for i in range(1, n_bins+1)] return bins, bin_means_y