weixin_44856491的博客

import numpy as npdef sigmoid(scores): return 1/(1+np.exp(-scores))class LogisticRegression: def fit(self,X,y,N=1000): m,n=X.shape w=np.zeros((n,1)) for t in range(N): pred=sigmoid(X.dot(w)) g=1.0

# logistic回归问题的梯度下降算法import numpy as npdef sigmoid(scores): #实现sigmoid函数 return 1/(1+np.exp(-scores))# 实现逻辑回归类class LogisticRegression: # 学习速率eta,搜索轮数N def fit(self,X,y,eta=0.1,N=1000): m,n=X.shape w=np.zeros((n,1))

#线性回归问题的随机梯度下降import numpy as npclass LinearRegression: def fit(self,X,y,eta_0=10,eta_1=50,N=3000): m,n=X.shape w=np.zeros((n,1)) self.w=w for t in range(N): i=np.random.randint(m) x=X[i].resha

# 线性回归问题的小批量随机梯度下降算法import numpy as npclass LinearRegression: def fit(self,X,y,eta_0=10,eta_1=50,N=3000,B=10): m,n=X.shape w=np.zeros((n,1)) self.w=w for t in range(N): batch=np.random.randint(low=0,high=m

# 线性回归问题的梯度下降算法import numpy as npclass LinearRegression: def fit(self,X,y,eta,N): m,n=X.shape w=np.zeros((n,1)) for t in range(N): e=X.dot(w)-y g=2*X.T.dot(e)/m w=w-eta*g self.w=w

import numpy as np# 线性回归模型class LinearRegression: def fit(self,X,y): # 计算最优解w=((X.T*X)^-1)*X.T*y self.w=np.linalg.inv(X.T.dot(X)).dot(X.T).dot(y) def predict(self,X): return X.dot(self.w)# 均方误差函数def mean_squared_error

# Lasso回归问题的坐标下降算法import numpy as npclass Lasso: def __init__(self,Lambda=1): self.Lambda=Lambda def soft_threshold(self,t,x): if x>t: return x-t elif x>=-t: return 0 else:

# lasso回归问题的次梯度下降算法import numpy as npclass Lasso: def __init__(self,Lambda=1): self.Lambda=Lambda def fit(self,X,y,eta=0.1,N=1000): m,n=X.shape w=np.zeros((n,1)) self.w=w for t in range(N): e=

import numpy as npfrom sklearn import datasetsfrom sklearn.model_selection import train_test_splitimport matplotlib.pyplot as pltfrom machine_learning.lib.kernel_svm import KernelSVMdef rbf_kernel(x1,x2): sigma=1.0 return np.exp(-np.linalg.no

import numpy as npfrom machine_learning.lib.svm_smo import SVMclass KernelSVM(SVM): def __init__(self,kernel=None): self.kernel=kernel def get_K(self,X_1,X_2): if self.kernel==None: return X_1.dot(X_2.T) m1,m

import numpy as npimport pandas as pdfrom sklearn.model_selection import train_test_splitfrom machine_learning.lib.decision_tree_regressor import DecisionTreeRegressorfrom sklearn.metrics import r2_scoredef get_data(): df=pd.read_csv("./bike.csv"

import numpy as npfrom machine_learning.lib.decision_tree_base import DecisionTreeBase# 对下标属于集合idx的数据计算标签的方差def get_var(y,idx): y_avg=np.average(y[idx])*np.ones(len(idx)) return np.linalg.norm(y_avg-y[idx],2)**2/len(idx)#定义决策树回归算法DecisionTreeRe

import numpy as npfrom machine_learning.lib.tree_node import Node'''类决策树基类的实现，类其中有4个成员：root，max_depth,get_score,feature_sample_rate其中：root记录所生成决策树的根节点，由根节点可唯一的表示这颗决策树max_depth 记录CART算法中决策树模型的深度限制，有构造函数指定get_score是分值函数的指针，由构造函数传入。传入方差则实现决策树回归问题的CART算法，