MatLab中的多变量梯度下降可预测房屋价格

Question

我开始编写程序，使用多变量梯度下降法基于5个要素来预测房价。我现在被困住了。我要遵循哪些步骤在MatLab中实现多变量梯度下降？我知道这听起来很愚蠢，但是我真的在这里迷路了，但我真的很想学习。如果有人帮助我，那将是一件非常伟大的事情。 到目前为止，这是我的代码：-

      %% 
% Converting Excel Data to a Matrix
Data= readmatrix("Flatdata.xlsx")
%%
% Seperating Data into its Columns

Year= Data(:,1);
Floor= Data(:,2);
Bedrooms= Data(:,3);
Area= Data(:,4);
PriceInLakhs= Data(:,5);
%% 
% Adjusting the Price for inflation

CPIInitial= readmatrix("CPI Values.xlsx");
A= abs(2000-Year);
A= A+1;
B= CPIInitial(A,2);
InflatedPriceInLakhs= PriceInLakhs.* (126.24./B);
Data(:,5)= InflatedPriceInLakhs;
%% 
% m= no. of training examples, n= no. of features

[m n]= size(Data);
n= n - 1;
%% 
% Feature Scaling

Year= (Year - mean(Year))./(max(Year) - min(Year));

Area= (Area - mean(Area))./(max(Area) - min(Area));
%% 
% Seperating Features and Prices

X = [Year Floor Bedrooms Area]

y = InflatedPriceInLakhs;
y= (y - mean(y))./(max(y) - min(y))
%% 
%