Linear regression with gradient descent to predict house prices using area one var in python

0 votes

I am trying to train a model using linear regression and gradient descent to predict the house prices based on area in sq ft but idk what is  wrong, the predicted hx (hypothesis) is larger than values. have not coded the for the prediction as i am getting error. help me out. ty

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

db = pd.read_csv("train.csv")
x = np.array(db.iloc[:40,5])
y = np.array(db.iloc[:40,11])
m = len(x)
t0=0
t1=1
alpha = 0.001
for i in range(m):
    hx = t0 + t1*x
    dt0 = (-2/m)*sum((hx - y))
    dt1 = (-2/m)*sum(x*(hx - y))
    t0 = t0 - dt0*alpha
    t1 = t1 - dt1*alpha
print(t0,t1)
plt.scatter(x,y)
plt.plot(x,hx)

also dataset,

x = [1300.236407 , 1275. , 933.1597222, 929.9211427, 999.009247 , 1250. , 1495.053957 , 1181.012946 , 1040. , 879.1208791, 1350.308642 , 1333.010179 , 927.1779023, 1122.171946 , 649.9837504, 1394.117647 , 1800.08471 , 2124.896706 , 1100. , 2178.649237 , 881.1435285, 944.8818898, 1310.147689 , 630.00063 , 1219.80971 , 780.141844 , 1600. , 1180.412371 , 1000. , 1000. , 1400.107701 , 943.1266076, 1150.146382 , 864.0674394, 857.7861968, 1174.210077 , 1020.087884 , 1650.165017 , 1000. , 1300.052002 ]

y = [ 55. , 51. , 43. , 62.5, 60.5, 42. , 66.5, 52. , 41.6, 36. , 35. , 110. , 48. , 62. , 20. , 71.1, 85. , 180. , 22. , 120. , 45. , 42. , 55. , 300. , 50. , 27.5, 46. , 22.9, 39. , 12.5, 52. , 33. , 55. , 82. , 240. , 55. , 65. , 65. , 35. , 75. ]

Mar 21 in Machine Learning by Dev
• 6,000 points
132 views

1 answer to this question.

0 votes
Apart from instructional purposes, I'm not sure why you require gradient descent for this and why you need to develop an algorithm from the start. For this, I propose using well-known tools such as Sklearn or Statsmodels.

Otherwise, linear regression can be performed directly like this, or you can utilize the aforementioned programs to choose which solver to use if you wish to use logistic (a type of linear regression) (I am not sure whether gradient descent is one of them as there are more controllable ways to solve linear problems that that).
answered Mar 23 by Nandini
• 5,480 points

Related Questions In Machine Learning

0 votes
1 answer

Why do we use gradient descent in linear regression?

An example you gave is one-dimensional, which ...READ MORE

answered Mar 23 in Machine Learning by Dev
• 6,000 points
137 views
0 votes
2 answers
+1 vote
2 answers

how can i count the items in a list?

Syntax :            list. count(value) Code: colors = ['red', 'green', ...READ MORE

answered Jul 7, 2019 in Python by Neha
• 330 points

edited Jul 8, 2019 by Kalgi 2,828 views
0 votes
1 answer
0 votes
1 answer
0 votes
1 answer
webinar REGISTER FOR FREE WEBINAR X
Send OTP
REGISTER NOW
webinar_success Thank you for registering Join Edureka Meetup community for 100+ Free Webinars each month JOIN MEETUP GROUP