Question

PYTHON

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.linear_model import LinearRegression
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split

Our goal is to create a linear regression model to estimate values of ln_price using ln_carat as the only feature. We will now prepare the feature and label arrays.

Our goal is to create a linear regression model to estimate values of In_price using In_carat as the only feature. We will now prepare the feature and label arrays. Perform the following steps in a single code cell: Create a 2D feature array named X2 containing the relevant feature, las well as a 1D label array named y2 containing the labels. (Note: These should be arrays, and not DataFrames or Series.) Use train_test_split() to split the data into training and testing sets using an 90/10 split. Name the resulting arrays X_train_2,x_test_2, y_train_2, and y_test_2. Set random state=1. Print the shapes of X_train_2 and X_test_2. Include text labeling the two results as shown below. Add spacing to ensure that the shape tuples are left-aligned. Training Features Shape: XXXX Test Features Shape: XXXX We will now create a linear regression model that can be used to estimate the natural log of the price of a diamond. Create a linear regression model named dia_mod and then fit it to the training data. Display the intercepts and coefficients for the final model with text labels as shown below. Add spacing to ensure that the values replacing the Xxxx characters are left-aligned. The intercept should appear as a single number, and the coefficients should be in the form of an array and should be displayed on a single line. Intercept: XXXX Coefficients: XXXX

"carat"   "cut" "color"   "clarity"   "depth"   "table"   "price"   "x"   "y"   "z"
"1" 0.23   "Ideal" "E" "SI2" 61.5 55 326   3.95   3.98   2.43
"2" 0.21   "Premium" "E" "SI1" 59.8 61 326   3.89   3.84   2.31
"3"   0.23   "Good" "E" "VS1" 56.9 65 327   4.05   4.07   2.31
"4"   0.29   "Premium" "I" "VS2" 62.4 58 334   4.2 4.23   2.63
"5"   0.31   "Good" "J" "SI2" 63.3 58 335   4.34   4.35   2.75
"6"   0.24   "Very Good"   "J" "VVS2"   62.8 57 336   3.94   3.96   2.48
"7"   0.24   "Very Good"   "I" "VVS1"   62.3 57 336   3.95   3.98   2.47
"8"   0.26   "Very Good"   "H" "SI1" 61.9 55 337   4.07   4.11   2.53
"9"   0.22 "Fair" "E" "VS2" 65.1 61 337   3.87 3.78   2.49
"10"   0.23   "Very Good"   "H"   "VS1" 59.4 61 338   4 4.05 2.39

Answer 1

I have written comments for the code. Please go through them

The data points given are less so for splitting i have used 0.2 percentage please change that 0.1 while you are using the script i have included the line as comment just uncomment it and use it when you have enough data. I have mentioned in the code as well please go through that.

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.linear_model import LinearRegression
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split

df = pd.read_csv('data.csv') # Read as csv file 

X2 = np.log(df['carat'].to_numpy().reshape(df.shape[0],1)) # converting carat to ln_carat and numpy  2d array
y = np.log(df['price'].to_numpy()) # converting price to ln_price and numpy 1d array

# I have used 0.2 test size here because there are only 10 data points are given and to calculate r-squared error 
# we need atleast two data points in test data set. So please delete this line and use the commented line when you have enough data
X_train_2, X_test_2, y_train_2, y_test_2 = train_test_split(X2, y, test_size=0.2, random_state = 1)
# X_train_2, X_test_2, y_train_2, y_test_2 = train_test_split(X2, y, test_size=0.1, random_state = 1) 

print("Training Features Shape: {}".format(X_train_2.shape))
print("Test Features Shape:     {}".format(X_test_2.shape))

dia_mod = LinearRegression() # Initalizing linear regression Model
dia_mod.fit(X_train_2, y_train_2) # fitting data with linear regression model.

print("Intercept:    {}".format(dia_mod.intercept_))
print("Coefficients: {}".format(dia_mod.coef_))


print('Training r-Squared: {}'.format(dia_mod.score(X_train_2,y_train_2)))
print('Testing r-Squared:  {}'.format(dia_mod.score(X_test_2,y_test_2)))

test_pred_2 = dia_mod.predict(X_test_2)
print("Observed Prices:  {}".format(np.exp(y_test_2))) # converting natural log values using exponential exp(log_base_e(2)) = 2
print("Estimated Prices: {}".format(np.exp(test_pred_2).round())) # converting natural log values using exponential exp(log_base_e(2)) = 2

diamonds_new = [0.5, 1.0, 1.5, 2.0, 2.5, 3.0] # given new list
diamonds_new = np.array(diamonds_new).reshape(len(diamonds_new),1) # conver to 2d column array 

new_pred_2 = dia_mod.predict(diamonds_new) # predicting using model
new_pred_3 = np.exp(new_pred_2) # converting natural log to normal values using np.exp
print("Esitmated Prices: {}".format(new_pred_3.round())) # round to nearest whole number/ prices.

Output Snippets:

-3]: import numpy as np import pandas as pd import matplotlib.pyplot as plt from sklearn.linear_model import LinearRegression from sklearn.linear_model import LogisticRegression from sklearn.model_selection import train_test_split 9]: df = pd.read_csv('data.csv') # Read as csv file 1]: X2 = np.log(df['carat'].to_numpy().reshape(df.shape[o],1)) # converting carat to Ln_carat and numpy 2d array #2]: y = np.log(df['price'].to_numpy()) # converting price to Ln_price and numpy ld array 3]: # I have used 0.2 test size here because there are only 10 data points are given and to calculate r-squared error # we need atleast two data points in test data set. so please delete this line and use the commented line when you have enough do X_train_2, X_test_2, y_train_2, y_test_2 = train_test_split(x2, y, test_size=0.2, random_state = 1) # X_train_2, x_test_2, y_train_2, y_test_2 = train_test_split(x2, y, test_size=0.1, random_state = 1)

: print("Training Features Shape: {}".format(X_train_2. shape)) print("Test Features Shape: {}".format(X_test_2. shape)) Training Features Shape: (8, 1) Test Features Shape: (2, 1)

1: dia_mod LinearRegression () # Initalizing Linear regression Model : dia_mod. fit(X_train_2, y_train_2) # fitting data with Linear regression model. |: LinearRegression() : print("Intercept: {}".format(dia_mod.intercept_)) print("Coefficients: {}".Format(dia_mod.coef_)) Intercept: 5.869718785239034 Coefficients: [0.94341585) : print('Training r-Squared: {}'.format(dia_mod.score (X_train_2,y_train_2))) print('Testing r-Squared: {}'.format(dia_mod.score (X_test_2,y_test_2))) Training r-squared: 0.17196034153466755 Testing r-squared: -0.0012783926126898049 : test_pred_2 = dia_mod. predict(X_test_2) print("Observed Prices: {}".Format(np.exp(y_test_2))) # converting natural log values using exponential exp(Log_base_e(2)) = 2 print("Estimated Prices: {}".Format(np.exp(test_pred_2).round())) # converting natural log values using exponential exp(Log_base_ Observed Prices: (327. 338. ] Estimated Prices: [332. 332.]

Please drop a comment for any further doubts.

- : diamonds_new [0.5, 1.0, 1.5, 2.0, 2.5, 3.0] # given new list diamonds_new = np.array(diamonds_new).reshape(len (diamonds_new),1) # conver to 2d column array : new_pred_2 dia_mod. predict(diamonds_new) # predicting using model new_pred_3 = np.exp(new_pred_2) # converting natural log to normal values using np.exp : print("Esitmated Prices: {}".format(new_pred_3.round())) # round to nearest whole number/ prices. Esitmated Prices: [362. 370. 378. 386. 395. 403. ]