Problem $1.$ Boston house data$.$

$$ CRIM, RM$,$ LSTAT, AGE $.(50)$

$(1),.$ Google drive$$

$$hw$1\_$bostonhouse.ipynb$.(30)$

$(2)$ Cost function graph$.(10)$

$(3)$ Scikit learn$$ Linear regression library$$

$.(10)$

# HW$1.$ House price prediction using linear regression

# import library

import tensorflow as tf

import numpy as np

import pandas as pd

from sklearn.model$\_$selection import train$\_$test$\_$split

# X$,$ Y

boston $=$ pd$.$read$\_$csv$(\text{'}$Boston$\_$house.csv$\text{'})$

X $=$ boston$[[\text{'}$CRIM$\text{'},\text{'}$RM$\text{'},$ 'LSTAT', 'AGE'$]]$

y $=$ boston$[[\text{'}$Target$\text{'}]]$

X$\_$train, X$\_$test, y$\_$train, y$\_$test $=$ train$\_$test$\_$split$($X$,$ y$,$ test$\_$size$=0\mathrm{.}2,$ random$\_$state$=42)$

X$\_$train $=$ np$.$array$($X$\_$train, dtype$=$np$.$float$32)$

X$\_$test $=$ np$.$array$($X$\_$test, dtype$=$np$.$float$32)$

y$\_$train $=$ np$.$array$($y$\_$train, dtype$=$np$.$float$32)$

y$\_$test $=$ np$.$array$($y$\_$test, dtype$=$np$.$float$32)$

# Model parameter

# $.$

# learning rate

# $$ learning rate$.$

learning$\_$rate $=0\mathrm{.}1$

# Gradient descent

def gradientDescent$()$:

with tf$.$GradientTape$()$ as tape:

# $($Gradient descent$).$

cost $=[]$

for step in range$(10001)$:

gradientDescent$()$

pred $=$ tf$.$matmul$($X$\_$test,W$)+$ b

cost$\_$inst $=$ tf$.$reduce$\_$mean$($tf$.$square$($pred $-$ y$\_$test$))$

cost $=$ np$.$append$($cost$,$ cost$\_$inst$)$