随机森林_处理不均衡数据

随机森林_处理不均衡数据

balanced 加上balanced 参数

# 处理不均衡的数据
from sklearn.ensemble import RandomForestClassifier
from sklearn import datasets
from sklearn.feature_selection import SelectFromModel
​
iris &＃61; datasets.load_iris()
features &＃61; iris.data
target &＃61; iris.target
# 删除前40个
features &＃61; features[40:, :]
target &＃61; target[40:]
# 二值化
target &＃61; np.where((target &＃61;&＃61; 0), 0, 1)
# balanced 加上balanced 参数
randomforest &＃61; RandomForestClassifier(random_state&＃61;0, n_jobs&＃61;-1, class_weight&＃61;"balanced")
# 训练模型 可设置权重值
model &＃61; randomforest.fit(features, target)
Discussion
A useful argument is balanced, wherein classes are automatically weighted inversely proptional to how frequently they appear in the data:
wj&＃61;nknj
wj&＃61;nknjwhere wjwj is the weight to class j, n is the number of observations, njnj is the number of observations in class j, and k is the total number of classes.