机器学习笔记

Posted 2023-02-15 花落风雪

大数据

• 机器学习

机器学习

回归问题和分类问题

• 导入数学函数库import numpy as np
• 导入绘图模块import matplotlib.pyplot as plt
• 生成回归样本数据from sklearn.datasets import make_regression
• X,y=make_regression(n_samples=100,n_features=1,n_informative=1,noise=50,random_state=8)

英文	翻译
n_samples	样本数
n_features	特征数(自变量个数)
n_informative	参与建模特征数
n_targets	因变量个数
noise	噪音
bias	偏差(截距)
coef	是否输出coef标识
random_state	随机状态若为固定值则每次产生的数据都一样

• 读取矩阵的长度X.shape和y.shape
• 构建画布fig=plt.figure()
• 引入坐标系ax=fig.add_subplot(111)
• 带入点ax. Scatter(X,y,c='b',s=60)
• 保存画布plt.savefig('lr.pdf')
• 导入y=ax+b的模型from sklearn.linear_model import LinearRegression
• 建立模型对象,构造函数lr=LinearRegression()
• 训练数据lr.fit(X,y)
• 求斜率lr.coef_
• 求截距lr.intercept_
• XX=np.linspace(-3,3,200)
  array = numpy.linspace(start, end, num=num_points) 将在 start 和 end 之间生成
  一个统一的序列，共有 num_points 个元素。
• XX=XX.reshape(-1,1)
• XX.shape
• yy=lr.predict(XX)
• ax.plot(XX,yy,c=‘r’)
• plt.savefig(‘test.pdf’)
• lr.score(X,y)
• lr.score(XX,yy)
• del X,XX,y,yy,fig,ax
• from sklearn.datasets import load_diabetes
• diabetes=load_diabetes()
  diabetes
• print(diabetes[‘DESCR’])
• from sklearn.model_selection import train_test_split
• X_train,X_test,y_train,y_test=train_test_split(X,y,random_state=8)
• X_train.shape
• X_test.shape
• lr=LinearRegression()
• lr.fit(X_train,y_train)
• lr.coef_
• lr.intercept_
• lr.score(X_train,y_train)
• lr.score(X_test,y_test)
  下节课：训练的分数跟测试分数差别很大，产生过拟合问题（岭回归和套索回归）

import numpy as np
from sklearn.datasets import load_diabetes
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from sklearn.linear_model import Lasso
from sklearn.linear_model import Ridge

diabetes=load_diabetes()
X=diabetes['data']
y=diabetes['target']
X_train,X_test,y_train,y_test=train_test_split(X,y,random_state=8)
lr=LinearRegression()
lr.fit(X_train,y_train)
print(lr.score(X_train,y_train))
print(lr.score(X_test,y_test))
rd=Ridge()
rd.fit(X_train,y_train)
print(rd.score(X_train,y_train))
print(rd.score(X_test,y_test))
la=Lasso()
la.fit(X_train,y_train)
print(la.score(X_train,y_train))
print(la.score(X_test,y_test))