Practice Exercises

Practice exercise from Chapter 2: Business Analytics Foundations

from sklearn.datasets import fetch_california_housing

california_housing = fetch_california_housing()

X = california_housing.data
y = california_housing.target

print(X)
print(y)

Practice exercise from Chapter 2: Business Analytics Foundations

columns = california_housing.feature_names
print(columns)

target = california_housing.target_names
print(target)

Practice exercise from Chapter 2: Business Analytics Foundations

import pandas as pd

california_df = pd.DataFrame(data = X, columns= columns)

california_df.head(10)

california_df[&

california_df[&

california_df[&

california_df.head(10)

Practice exercise from Chapter 2: Business Analytics Foundations

california_df.shape

Practice exercise from Chapter 2: Business Analytics Foundations

california_df.dtypes

california_df[&
california_df[&

california_df.dtypes

Practice exercise from Chapter 4: Data Collection and Preparation

california_df.isnull()

california_df.isnull().sum()

california_df.isna()

california_df.isna().sum()

import numpy as np

def introduce_missing_values(df, fraction_missing, random_state=None):
    np.random.seed(random_state)
    mask = np.random.rand(*df.shape) < fraction_missing
    df_with_missing = df.mask(mask)
    return df_with_missing

df_with_missing = introduce_missing_values(california_df, 0.2, random_state=42)

print("Original DataFrame:")
print(california_df.head())

print("\nDataFrame with random missing values:")
print(df_with_missing.head())

df_with_missing.head(10)

df_with_missing.isna().sum()

import pandas as pd
import matplotlib.pyplot as plt

missing_proportion = df_with_missing.isnull().mean()

total_rows = len(df_with_missing)

missing_count = df_with_missing.isnull().sum()

missing_count_df = pd.DataFrame({&

plt.figure(figsize=(10, 6))
plt.bar(missing_proportion.index, total_rows, color=&
plt.bar(missing_proportion.index, missing_count_df[&
plt.title(&
plt.xlabel(&
plt.ylabel(&
plt.xticks(rotation=45)
plt.legend([&
plt.tight_layout()
plt.show()

df_with_missing.dropna(subset=[&
print(df_with_missing.isna().sum())
print(df_with_missing.shape)

df_with_missing.drop(&
print(df_with_missing.isna().sum())
print(df_with_missing.head(10))

lat_mode = df_with_missing[&
print(lat_mode)
df_with_missing[&
print(df_with_missing.isna().sum())

df_with_missing[&
print(df_with_missing[&
print(df_with_missing[&
print(df_with_missing[&

mean_houseage = df_with_missing[&
print(mean_houseage)
df_with_missing[&
print(df_with_missing.isna().sum())
print(df_with_missing[&

df_with_missing[&

Practice exercise from Chapter 4: Data Collection and Preparation

california_df.boxplot()

plt.xticks(rotation=45, ha=&

plt.show()

def count_outliers(column):
    if pd.api.types.is_numeric_dtype(column):
        Q1 = column.quantile(0.25)
        Q3 = column.quantile(0.75)
        IQR = Q3 - Q1

        lower_bound = Q1 - 1.5 * IQR
        upper_bound = Q3 + 1.5 * IQR

        outliers = column[(column < lower_bound) | (column > upper_bound)]
        return outliers.shape[0]
    else:
        return 0

numeric_columns = california_df.select_dtypes(include=[&

outliers_count = numeric_columns.apply(count_outliers)

print("Number of outliers in each numeric column:")
print(outliers_count)

def remove_outliers(column):
    Q1 = column.quantile(0.25)
    Q3 = column.quantile(0.75)
    IQR = Q3 - Q1

    lower_bound = Q1 - 1.5 * IQR
    upper_bound = Q3 + 1.5 * IQR

    column = column[(column >= lower_bound) & (column <= upper_bound)]

    return column

california_df_clean = california_df.copy()

for column in california_df_clean.select_dtypes(include=[&
    california_df_clean[column] = remove_outliers(california_df_clean[column])

california_df_clean = california_df_clean.dropna()

print("
DataFrame after removing outliers:")
print(california_df_clean)

import matplotlib.pyplot as plt

california_df_clean.boxplot(figsize=(10, 6))

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.xticks(rotation=45)
plt.tight_layout()
plt.show()

california_df_copy = california_df.copy()

california_df_copy.boxplot()

plt.xticks(rotation=45, ha=&

plt.show()

california_df_copy = california_df_copy[california_df_copy[&

california_df_copy.boxplot()

plt.xticks(rotation=45, ha=&

plt.show()

california_df_copy.shape

Practice exercise from Chapter 4: Data Collection and Preparation

import pandas as pd
from sklearn.preprocessing import StandardScaler

scaler = StandardScaler()

numeric_columns = california_df.select_dtypes(include=&

california_df_standardized = scaler.fit_transform(numeric_columns)

california_df_standardized = pd.DataFrame(california_df_standardized, columns=numeric_columns.columns)

print("Standardized California DataFrame:")
print(california_df_standardized)

california_df_standardized.boxplot()

plt.xticks(rotation=45, ha=&

plt.show()

Practice exercise from Chapter 4: Data Collection and Preparation

from sklearn.preprocessing import MinMaxScaler

scaler = MinMaxScaler()

numeric_columns = california_df.select_dtypes(include=&

california_df_scaled = scaler.fit_transform(numeric_columns)

california_df_scaled = pd.DataFrame(california_df_scaled, columns=numeric_columns.columns)

print("Scaled California DataFrame:")
print(california_df_scaled)

california_df_scaled.boxplot()

plt.xticks(rotation=45, ha=&

plt.show()

X = california_df_scaled.drop([&
y = california_df_scaled[&

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 42)

model = LinearRegression()

model.fit(X_train, y_train)

y_pred = model.predict(X_test)

mse = mean_squared_error(y_pred, y_test)
print(mse)
r2 = r2_score(y_test, y_pred)
print(r2)

Practice exercise from Chapter 7: Data Visualization

california_df[&

import matplotlib.pyplot as plt

california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&
plt.show()

california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

california_df.hist(figsize=(15, 10), color=&
plt.suptitle(&
plt.show()

Practice exercise from Chapter 7: Data Visualization

california_df.boxplot(&

import matplotlib.pyplot as plt

plt.figure(figsize=(8, 6))
plt.boxplot(california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

import matplotlib.pyplot as plt

plt.figure(figsize=(8, 6))
plt.boxplot(california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

import matplotlib.pyplot as plt

plt.figure(figsize=(8, 6))
plt.boxplot(california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

import matplotlib.pyplot as plt

plt.figure(figsize=(8, 6))
plt.boxplot(california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

import matplotlib.pyplot as plt

plt.figure(figsize=(8, 6))
plt.boxplot(california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

import matplotlib.pyplot as plt

plt.figure(figsize=(8, 6))
plt.boxplot(california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

import matplotlib.pyplot as plt

plt.figure(figsize=(8, 6))
plt.boxplot(california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

import matplotlib.pyplot as plt

numeric_columns = california_df.select_dtypes(include=[&

plt.figure(figsize=(12, 8))
plt.boxplot(california_df[numeric_columns].values, patch_artist=True)
plt.xticks(range(1, len(numeric_columns) + 1), numeric_columns, rotation=45)
plt.title(&
plt.xlabel(&
plt.ylabel(&
plt.tight_layout()

plt.show()

latitude_values = california_df[&

latitude_values.head(10)

latitude_values.shape

top_50_categories = latitude_values.head(50)

plt.figure(figsize=(20, 6))
top_50_categories.plot(kind=&

plt.title(&
plt.xlabel(&
plt.ylabel(&
plt.xticks(rotation=45)

plt.show()

import matplotlib.pyplot as plt

top_50_categories = california_df[&

plt.figure(figsize=(20, 6))
plt.bar(range(len(top_50_categories)), top_50_categories.values, color=&

plt.title(&
plt.xlabel(&
plt.ylabel(&
plt.xticks(range(len(top_50_categories)), top_50_categories.index, rotation=45)

plt.tight_layout()
plt.show()

top_50_latitude = california_df[&
top_50_longitude = california_df[&

plt.figure(figsize=(20, 6))

plt.subplot(1, 2, 1)
plt.bar(range(len(top_50_latitude)), top_50_latitude.values, color=&
plt.title(&
plt.xlabel(&
plt.ylabel(&
plt.xticks(range(len(top_50_latitude)), top_50_latitude.index, rotation=45)

plt.subplot(1, 2, 2)
plt.bar(range(len(top_50_longitude)), top_50_longitude.values, color=&
plt.title(&
plt.xlabel(&
plt.ylabel(&
plt.xticks(range(len(top_50_longitude)), top_50_longitude.index, rotation=45)

plt.tight_layout()
plt.show()

Practice exercise from Chapter 7: Data Visualization

california_df.plot.scatter(x=&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

plt.figure(figsize=(10, 6))
plt.scatter(california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

california_df.dtypes

plt.figure(figsize=(10, 6))
plt.scatter(california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

plt.figure(figsize=(10, 6))
plt.scatter(california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

plt.figure(figsize=(10, 6))
plt.scatter(california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

plt.figure(figsize=(10, 6))
plt.scatter(california_df[&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

Practice exercise from Chapter 7: Data Visualization

california_df.corr(numeric_only=True)

import seaborn as sns

correlation_matrix = california_df.corr(numeric_only=True)

plt.figure(figsize=(10, 8))
sns.heatmap(correlation_matrix, annot=True, cmap=&
plt.title(&
plt.show()

Practice exercise from Chapter 7: Data Visualization

sns.pairplot(california_df)
plt.show()

Practice exercise from Chapter 7: Data Visualization

import matplotlib.pyplot as plt
import numpy as np

houseage_min = california_df[&
houseage_max = california_df[&
normalized_houseage = (california_df[&

top_70_categories = california_df.groupby(&

plt.figure(figsize=(20, 6))
bars = top_70_categories.plot(kind=&

cbar = plt.colorbar(plt.cm.ScalarMappable(cmap=&
cbar.set_label(&

legend_labels = [f&
cbar.set_ticks([0, 1])
cbar.set_ticklabels(legend_labels)

plt.title(&
plt.xlabel(&
plt.ylabel(&
plt.xticks(rotation=45)

plt.show()

import matplotlib.pyplot as plt

scatter = plt.scatter(x=&

plt.colorbar(scatter, label=&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

import matplotlib.pyplot as plt

scatter = plt.scatter(x=&

plt.colorbar(scatter, label=&

plt.title(&
plt.xlabel(&
plt.ylabel(&

plt.show()

Practice exercise from Chapter 8: Statistical Analysis

california_df[&

california_df[&

california_df[&

california_df.describe(include=&

california_df.describe(include=&

california_df.describe(include=&

california_df.describe()

Practice exercise from Chapter 8: Statistical Analysis

from sklearn.linear_model import LinearRegression
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error, r2_score

X = california_df[&
y = california_df[&

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 42)

X_train_reshaped = X_train.values.reshape(-1, 1)
X_test_reshaped = X_test.values.reshape(-1, 1)

model = LinearRegression()

model.fit(X_train_reshaped, y_train)

model.intercept_

model.coef_

y_pred = model.predict(X_test_reshaped)

mse = mean_squared_error(y_pred, y_test)
print(mse)
r2 = r2_score(y_test, y_pred)
print(r2)

Practice exercise from Chapter 8: Statistical Analysis

X = california_df.drop([&
y = california_df[&

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 42)

model = LinearRegression()

model.fit(X_train, y_train)

model.intercept_

model.coef_

y_pred = model.predict(X_test)

mse = mean_squared_error(y_pred, y_test)
print(mse)
r2 = r2_score(y_test, y_pred)
print(r2)

Practice exercise from Chapter 8: Statistical Analysis

coef_names = california_df.drop(columns=[&

for name, coef in zip(coef_names, model.coef_):
    print(f"{name}: {coef}")

print("Intercept (model intercept):", model.intercept_)

equation = "y = "
for name, coef in zip(coef_names, model.coef_):
    equation += f"{coef:.4f} * {name} + "
equation += f"{model.intercept_:.4f}"
print("Regression equation:", equation)

sorted_coefs = sorted(zip(coef_names, model.coef_), key=lambda x: abs(x[1]), reverse=True)

for name, coef in sorted_coefs:
    print(f"{name}: {coef:.4f}")

import statsmodels.api as sm

X = sm.add_constant(california_df.drop(columns=[&

model = sm.OLS(california_df[&

p_values = model.pvalues

print("P-values for the coefficients:")
print(p_values)

Practice exercise from Chapter 8: Statistical Analysis

residuals = y_test - y_pred

residuals.hist()
plt.title(&
plt.xlabel(&
plt.ylabel(&
plt.show()

plt.figure(figsize=(8, 6))
plt.boxplot(residuals)
plt.title(&
plt.xlabel(&
plt.ylabel(&
plt.show()

residuals.describe()

Practice exercise from Chapter 8: Statistical Analysis

X = california_df_standardized.drop([&
y = california_df_standardized[&

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 42)

model = LinearRegression()

model.fit(X_train, y_train)

y_pred = model.predict(X_test)

mse = mean_squared_error(y_pred, y_test)
print(mse)
r2 = r2_score(y_test, y_pred)
print(r2)

X = california_df.drop([&
y = california_df[&

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 42)

model = LinearRegression()

model.fit(X_train, y_train)

y_pred = model.predict(X_test)

mse = mean_squared_error(y_pred, y_test)
print(mse)
r2 = r2_score(y_test, y_pred)
print(r2)

X = california_df_scaled.drop([&
y = california_df_scaled[&

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 42)

model = LinearRegression()

model.fit(X_train, y_train)

y_pred = model.predict(X_test)

mse = mean_squared_error(y_pred, y_test)
print(mse)
r2 = r2_score(y_test, y_pred)
print(r2)