[英]Showing ValueError: shapes (1,3) and (1,3) not aligned: 3 (dim 1) != 1 (dim 0)
[英]Least Square Implementation error "ValueError: shapes (3,3) and (1,3) not aligned: 3 (dim 1) != 1 (dim 0)"
嗨,我正在尝试为一个项目实现线性回归模块,但我卡在这里的是 git repo: 最小二乘法
编码 :
import numpy as np
from numpy.linalg import inv
class linearregression :
def __init__(self,training_data_X, training_data_Y) :
# Linear regression module init
pading = np.ones(training_data_X.shape[0])
self.training_data_X = np.insert(training_data_X, 0, pading, axis=1) # The training data x => features numpy_matrix
self.training_data_Y = training_data_Y # The training data y => response numpy_matrix
def leastsquare(self):
# Find beta parameter
X_transpose = self.training_data_X.transpose()
self.ajk = np.matmul(X_transpose, self.training_data_X)
self.Hessian = 2 * self.ajk
self.Beta = inv(self.ajk).dot(X_transpose).dot(self.training_data_Y)
f_of_X = self.training_data_X.dot(self.Beta)
self.rs = np.subtract(self.training_data_Y,f_of_X)
self.rss = self.rs.dot(self.rs)
return self.Beta , self.rss
if __name__ == "__main__" :
x = np.matrix([[0,1],[1,4],[7,8]])
y = np.matrix([2,9,23])
Lr = linearregression(x,y)
Beta, rss = Lr.leastsquare()
print(Beta,rr)
代码返回我这个错误语句:
Traceback (most recent call last):
File "<string>", line 27, in <module>
File "<string>", line 17, in leastsquare
ValueError: shapes (3,3) and (1,3) not aligned: 3 (dim 1) != 1 (dim 0)
Y
是 1x3,但应该是 3x1。 修复它的两种方法:
training_data_Y
为 nx1: y = np.matrix([[2],[9],[23]])
y = np.array([2,9,23])
Beta = inv(self.ajk).dot(X_transpose).dot(self.training_data_Y.T)
我找到了它的解决方案:
y = np.matrix([[2],[9],[23]])
和
self.rss = np.square(self.rs)
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.