逻辑回归_处理不均衡的数据

逻辑回归_处理不均衡的数据
class_weight=“balanced” 参数 根军样本出现的评论自动给 样本设置 权重

# 处理不均衡的数据
import numpy as np
from sklearn.linear_model import LogisticRegression
from sklearn import datasets
from sklearn.preprocessing import StandardScaler
​
iris = datasets.load_iris()
# 移走40个数据，使数据不均衡
features = iris.data[40:, :]
target = iris.target[40:]
target
array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2,
       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2])
# 打标签 
target = np.where((target == 0), 0, 1)
target
# 标准化数据
scaler = StandardScaler()
features_standardized = scaler.fit_transform(features)
# class_weight="balanced"  参数  根军样本出现的评论自动给 样本设置 权重
logistic_regression = LogisticRegression(random_state=0, class_weight="balanced")
model = logistic_regression.fit(features_standardized, target)