使用梯度下降实现线性回归

问题描述

使用梯度下降的线性回归

参考：Linear Regression Using Gradient Descent in 10 Lines of Code

数据集：

房价卧室数

编码环境：Google合作实验室

代码：

import numpy as np
import pandas as pd
import io
from google.colab import files 
uploaded = files.upload()
df = pd.read_csv(io.BytesIO(uploaded['datasets_46927_85203_data.csv'])) 
house_price=df.price
No_of_bedrooms=df.bedrooms
# Mean of house_price and No_of_bedrooms
mean_x= np.mean(house_price)
mean_y=np.mean(No_of_bedrooms)

#Total number of datapoints
n=len(house_price)
numerator = 0;
denominator = 0;
for i in range (n):
  numerator += (house_price[i] - mean_x) * (No_of_bedrooms[i] - mean_y)
  denominator += (house_price[i] - mean_x) ** 2
slope= numerator / denominator
constant = mean_y - (slope*mean_x)
def linear_regression(house_price[],y,m_current=slope,b_current=constant,epochs=1000,learning_rate=0.0001):
     N = float(len(No_of_bedrooms))
     for i in range(epochs):
          y_current = (m_current * house_price[i]) + b_current
          cost = sum([data**2 for data in (No_of_bedrooms[i]-y_current)]) / N
          m_gradient = -(2/N) * sum(house_price[i] * (No_of_bedrooms[i] - y_current))
          b_gradient = -(2/N) * sum(No_of_bedrooms[i] - y_current)
          m_current = m_current - (learning_rate * m_gradient)
          b_current = b_current - (learning_rate * b_gradient)
     return m_current,b_current,cost

错误：def linear_regression(house_price[],learning_rate=0.0001):

Error Statement: File "<ipython-input-6-864b867cbc7a>",line 31
    

def linear_regression(house_price[],learning_rate=0.0001):
                                     ^
SyntaxError: invalid Syntax

我不知道如何消除此错误。

此post与我的查询无关。

如果有人帮助我，那将很棒。谢谢。

解决方法

暂无找到可以解决该程序问题的有效方法，小编努力寻找整理中！

如果你已经找到好的解决方法，欢迎将解决方案带上本链接一起发送给小编。

小编邮箱:dio#foxmail.com (将#修改为@）