adaboosting多分类

import numpy as np

from sklearn.ensemble import AdaBoostClassifier,AdaBoostRegressor

from sklearn import datasets

from sklearn.model_selection import train_test_split

from sklearn import tree

X,y = datasets.load_iris(True)
X_train,X_test,y_train,y_test = train_test_split(X,y,test_size = 0.2,random_state = 1024)

# 多分类问题，不是+1、-1
# 类别0,1,2
# 参数，这个参数，选多大合适，试一试：模型选择
# n_estimators参数，通过代码，进行筛选合适的参数
ada = AdaBoostClassifier(n_estimators=3,algorithm='SAMME',learning_rate=1.0)
ada.fit(X_train,y_train)#算法，工作：从X_train---y_train寻找规律
y_ = ada.predict(X_test)
proba_ = ada.predict_proba(X_test)
accuracy = ada.score(X_test,y_test)
print('--------------------算法准确率：',accuracy)

计算第一棵树弱学习器权重，更新样本权重

learning_rate = 1.0
# 鸢尾花，三分类问题
np.unique(y)
num = 3

# 初始化样本(训练数据，X_train)，每个样本的权重都是一样的
w0 = np.full(shape = 120,fill_value=1/120)
# 误差
y1 = ada[0].predict(X_train)#y1弱学习器预测值，y_train真实
e1 = ((y1 != y_train)*w0).sum()
# 弱学习器的权重，扁