課程作業 提取碼：3szr
1、單元線性回歸
ex1.py
from matplotlib.colors import lognorm
from mpl_toolkits.mplot3d import axes3d
from computecost import
*from plotdata import
*print
('plotting data...'
)data=np.loadtxt(
'./data/ex1data1.txt'
,delimiter=
',')
#載入txt格式資料集 每一行以「，」分隔
x=data[:,
0]y=data[:,
1]m=y.size
plt.figure(0)
plot_data(x,y)
# input()
print
('running gradient descent...'
)x=np.c_[np.ones(m)
,x]theta=np.zeros(2)
iterations=
1500
alpha=
0.01
print
('initial cost: '
+str
(compute_cost(x,y,theta))+
'(this value should be about 32.07'
)theta,j_history =gradient_descent(x,y,theta,alpha,iterations)
print
('theta found by gradient descent:'
+str
(theta.reshape(2)
))plt.figure(0)
line1,
=plt.plot(x[:,
1],np.dot(x,theta)
,label=
'linear regression'
)plot_data(x[:,
1],y)plt.legend(handles=
[line1]
)input
('program paused. press enter to continue'
)predict1=np.dot(np.array([1
,3.5])
,theta)
print
('for population=35,000,we predict a profit of (this value should be about 4519.77)'
.format
(predict1*
10000))
predict2=np.dot(np.array([1
,7])
,theta)
print
('for population = 70,000, we predict a profit of (this value should be about 45342.45)'
.format
(predict2*
10000))
input
('program paused. press enter to continue'
)print
('visualizing j(theta_0,theta_1)...'
)theta0_vals=np.linspace(-10
,10,100
)theta1_vals=np.linspace(-1
,4,100
)j_vals=np.zeros(
(theta0_vals.shape[0]
,theta1_vals.shape[0]
))print
(theta0_vals.shape[0]
)print
(theta1_vals.shape[0]
)for i in
range(0
,theta0_vals.shape[0]
):for j in
range(0
,theta1_vals.shape[0]
):t=np.array(
[theta0_vals[i]
,theta1_vals[j]])
j_vals[i]
[j]=compute_cost(x,y,t)
j_vals=np.transpose(j_vals)
fig=plt.figure(1)
ax=axes3d(fig)
xs,ys=np.meshgrid(theta0_vals,theta1_vals)
plt.title(
"visualizing j(theta_0,theta_1)"
)ax.plot_su***ce(xs,ys,j_vals)
ax.set_xlabel(
'$\theta_0$'
,color=
'r')
ax.set_ylabel(
'$\theta_1$'
,color=
'r')
plt.show(
)plt.figure(2)
lvls=np.logspace(-2
,3,20
)plt.contourf(xs,ys,j_vals,
10,levels=lvls,normal=lognorm())
plt.plot(theta[0]
, theta[1]
, c=
'r', marker=
"x")
plt.show(
)
import matplotlib.pyplot as plt
defplot_data
(x,y)
: plt.scatter(x,y,marker=
'x',s=
50,c=
'r',alpha=
0.8)
plt.xlabel(
'population'
) plt.ylabel(
'profits'
) plt.show(
)
吳恩達機器學習筆記 第二週
h x 0 1 x1 2x2 3x3 4x 4h theta x theta 0 theta 1x 1 theta 2x 2 theta 3x 3 theta 4x 4 h x 0 1 x1 2 x 2 3 x3 4 x4 為了表示方便，我們定義乙個單純為了計算方便的特徵，也就是x0 1x 0 1 ...
吳恩達機器學習第五周測驗及程式設計作業
程式設計練習 neural network learning 答案 答案 答案 答案 ad分析 a 使用梯度檢驗來檢查反向傳播是否正確，正確。b 梯度檢驗要比反向傳播計算損失函式的梯度慢的多，錯誤。c 梯度檢驗對梯度下降演算法來說非常有用，錯誤。d 為了保證效率，在使用反向傳播演算法前關閉梯度檢驗，...
吳恩達機器學習 程式設計練習7
本練習的主題是k means clustering and principal component analysis，即k均值聚類演算法和主要成分分析。因此這篇文章也分兩部分來討論，根據作業檔案的步驟來分別對k均值聚類演算法和pca進行討論原理以及如何實現。k means clustering al...