import pandas as pd

import matplotlib.pyplot as plt

import seaborn as sns

# Load the dataset

data = pd.read_csv('Walmart Sales data.csv.csv') # Replace
'sales_data.csv' with your actual file path

# Display the first few rows of the dataset to understand its

structure
print(data.head())

Invoice ID Branch City Customer type Gender \

0 750-67-8428 A Yangon Member Female
1 226-31-3081 C Naypyitaw Normal Female
2 631-41-3108 A Yangon Normal Male
3 123-19-1176 A Yangon Member Male
4 373-73-7910 A Yangon Normal Male

Product line Unit price Quantity Tax 5% Total \

0 Health and beauty 74.69 7 26.1415 548.9715
1 Electronic accessories 15.28 5 3.8200 80.2200
2 Home and lifestyle 46.33 7 16.2155 340.5255
3 Health and beauty 58.22 8 23.2880 489.0480
4 Sports and travel 86.31 7 30.2085 634.3785

Date Time Payment cogs gross margin percentage

\
0 2019-01-05 13:08:00 Ewallet 522.83 4.761905

1 2019-03-08 10:29:00 Cash 76.40 4.761905

2 2019-03-03 13:23:00 Credit card 324.31 4.761905

3 2019-01-27 20:33:00 Ewallet 465.76 4.761905

4 2019-02-08 10:37:00 Ewallet 604.17 4.761905

gross income Rating

0 26.1415 9.1
1 3.8200 9.6
2 16.2155 7.4
3 23.2880 8.4
4 30.2085 5.3
Q1. How many distinct cities are present in the
dataset?
distinct_cities = data['City'].nunique()
print("Number of distinct cities:", distinct_cities)

Number of distinct cities: 3

Q2. In which city is each branch situated?

branch_city_mapping = data.groupby('Branch')['City'].unique()
print("Branches and their respective cities:")
for branch, city in branch_city_mapping.items():
print("Branch:", branch, "-> City:", city)

Branches and their respective cities:

Branch: A -> City: ['Yangon']
Branch: B -> City: ['Mandalay']
Branch: C -> City: ['Naypyitaw']

Visualizations

Distribution of sales across branches

plt.figure(figsize=(10, 6))
sns.countplot(x='Branch', data=data, palette='Set2')
plt.title('Distribution of Sales Across Branches')
plt.xlabel('Branch')
plt.ylabel('Number of Sales')
plt.show()
Distribution of customer types
plt.figure(figsize=(10, 6))
sns.countplot(x='Customer type', data=data, palette='Pastel1')
plt.title('Distribution of Customer Types')
plt.xlabel('Customer Type')
plt.ylabel('Number of Customers')
plt.show()
Gender distribution
plt.figure(figsize=(10, 6))
sns.countplot(x='Gender', data=data, palette='Dark2')
plt.title('Gender Distribution of Customers')
plt.xlabel('Gender')
plt.ylabel('Number of Customers')
plt.show()
Product line distribution
plt.figure(figsize=(12, 6))
sns.countplot(y='Product line', data=data, palette='Set3')
plt.title('Distribution of Product Lines')
plt.xlabel('Number of Sales')
plt.ylabel('Product Line')
plt.show()
Distribution of ratings
plt.figure(figsize=(10, 6))
sns.histplot(data['Rating'], bins=10, kde=True, color='skyblue')
plt.title('Distribution of Ratings')
plt.xlabel('Rating')
plt.ylabel('Frequency')
plt.show()
3. How many distinct product lines are there in
the dataset?
distinct_product_lines = data['Product line'].nunique()
print("Number of distinct product lines:", distinct_product_lines)

Number of distinct product lines: 6

4. What is the most common payment method?

most_common_payment_method = data['Payment'].mode()[0]
print("Most common payment method:", most_common_payment_method)

Most common payment method: Ewallet

5. What is the most selling product line?

most_selling_product_line = data['Product
line'].value_counts().idxmax()
print("Most selling product line:", most_selling_product_line)

Most selling product line: Fashion accessories

6. What is the total revenue by month?

# Convert 'date' column to datetime format
data['Date'] = pd.to_datetime(data['Date'])

# Extract month from the 'date' column

data['month'] = data['Date'].dt.month

# Calculate total revenue by month

total_revenue_by_month = data.groupby('month')['Total'].sum()
print("Total revenue by month:")
print(total_revenue_by_month)

Total revenue by month:

month
1 116291.868
2 97219.374
3 109455.507
Name: Total, dtype: float64

7. Which month recorded the highest Cost of

Goods Sold (COGS)?
# Convert 'date' column to datetime format
data['Date'] = pd.to_datetime(data['Date'])

# Ensure the column has been converted successfully

print(data['Date'].dtype)

# Calculate total revenue by month

total_revenue_by_month = data.groupby(data['Date'].dt.month)
['Total'].sum()
print("Total revenue by month:")
print(total_revenue_by_month)

# Find the month with the highest Cost of Goods Sold (COGS)
highest_cogs_month = data.groupby(data['Date'].dt.month)
['cogs'].sum().idxmax()
print("Month with the highest Cost of Goods Sold (COGS):",
highest_cogs_month)
datetime64[ns]
Total revenue by month:
Date
1 116291.868
2 97219.374
3 109455.507
Name: Total, dtype: float64
Month with the highest Cost of Goods Sold (COGS): 1

8. Which product line generated the highest

revenue?
highest_revenue_product_line = data.groupby('Product line')
['Total'].sum().idxmax()
print("Product line with the highest revenue:",
highest_revenue_product_line)

Product line with the highest revenue: Food and beverages

9. Which city has the highest revenue?

highest_revenue_city = data.groupby('City')['Total'].sum().idxmax()
print("City with the highest revenue:", highest_revenue_city)

City with the highest revenue: Naypyitaw

10. Which product line incurred the highest

VAT?
# 10. Which product line incurred the highest VAT?
highest_vat_product_line = data.groupby('Product line')['Tax
5%'].sum().idxmax()
print("Product line with the highest VAT:", highest_vat_product_line)

Product line with the highest VAT: Food and beverages

11. Retrieve each product line and add a column
product_category, indicating 'Good' or 'Bad,'

based on whether its sales are above the

average.
# 11. Retrieve each product line and add a column product_category,
indicating 'Good' or 'Bad',
# based on whether its sales are above the average.

average_quantity_sold = data['Quantity'].mean()

# Function to categorize sales

def categorize_sales(quantity):
if quantity > average_quantity_sold:
return 'Good'
else:
return 'Bad'

# Apply the function to create the product category column

data['product_category'] = data['Quantity'].apply(categorize_sales)

# Display the updated DataFrame with the new column

print(data[['Product line', 'Quantity', 'product_category']].head())

Product line Quantity product_category

0 Health and beauty 7 Good
1 Electronic accessories 5 Bad
2 Home and lifestyle 7 Good
3 Health and beauty 8 Good
4 Sports and travel 7 Good

12. Which branch sold more products than

average product sold?
# 12. Which branch sold more products than average product sold?
branch_product_counts = data.groupby('Branch')['Quantity'].sum()
branch_more_than_average = branch_product_counts[branch_product_counts
> average_sales].index.tolist()
print("Branch(es) with more products sold than the average:",
branch_more_than_average)
Branch(es) with more products sold than the average: ['A', 'B', 'C']

13. What is the most common product line by

gender?
# 13. What is the most common product line by gender?
common_product_line_by_gender = data.groupby(['Gender', 'Product
line']).size().idxmax()
print("Most common product line by gender:",
common_product_line_by_gender[1])

Most common product line by gender: Fashion accessories

14. What is the average rating of each product

line?
# 14. What is the average rating of each product line?
average_rating_by_product_line = data.groupby('Product line')
['Rating'].mean()
print("Average rating of each product line:")
print(average_rating_by_product_line)

Average rating of each product line:

Product line
Electronic accessories 6.924706
Fashion accessories 7.029213
Food and beverages 7.113218
Health and beauty 7.003289
Home and lifestyle 6.837500
Sports and travel 6.916265
Name: Rating, dtype: float64

15. Number of sales made in each time of the

day per weekday
# 15. Number of sales made in each time of the day per weekday
data['weekday'] = data['Date'].dt.weekday
sales_per_time_per_weekday = data.groupby(['weekday', 'Time']).size()
print("Number of sales made in each time of the day per weekday:")
print(sales_per_time_per_weekday)

Number of sales made in each time of the day per weekday:

weekday Time
0 10:00:00 1
10:02:00 1
10:05:00 1
10:11:00 1
10:23:00 2
..
6 20:33:00 1
20:37:00 1
20:38:00 1
20:46:00 1
20:51:00 1
Length: 914, dtype: int64

16. Identify the customer type that generates

the highest revenue.
# 16. Identify the customer type that generates the highest revenue.
highest_revenue_customer_type = data.groupby('Customer type')
['Total'].sum().idxmax()
print("Customer type that generates the highest revenue:",
highest_revenue_customer_type)

Customer type that generates the highest revenue: Member

17. Which city has the largest tax percent/ VAT

(Value Added Tax)?
# 17. Which city has the largest tax percent/ VAT (Value Added Tax)?
city_with_largest_vat_percent = data.groupby('City')['Tax
5%'].mean().idxmax()
print("City with the largest tax percent/ VAT:",
city_with_largest_vat_percent)

City with the largest tax percent/ VAT: Naypyitaw

18. Which customer type pays the most VAT?
# 18. Which customer type pays the most VAT?
customer_type_with_most_vat = data.groupby('Customer type')['Tax
5%'].sum().idxmax()
print("Customer type that pays the most VAT:",
customer_type_with_most_vat)

Customer type that pays the most VAT: Member

19. How many unique customer types does the

data have?
# 19. How many unique customer types does the data have?
unique_customer_types = data['Customer type'].nunique()
print("Number of unique customer types:", unique_customer_types)

Number of unique customer types: 2

20. How many unique payment methods does

the data have?
# 20. How many unique payment methods does the data have?
unique_payment_methods = data['Payment'].nunique()
print("Number of unique payment methods:", unique_payment_methods)

Number of unique payment methods: 3

21. Which is the most common customer type?

most_common_customer_type = data['Customer type'].mode()[0]
print("Most common customer type:", most_common_customer_type)

Most common customer type: Member

22. Which customer type buys the most?

most_buying_customer_type = data.groupby('Customer type')
['Quantity'].sum().idxmax()
print("Customer type that buys the most:", most_buying_customer_type)

Customer type that buys the most: Member

23. What is the gender of most of the

customers?
most_common_gender = data['Gender'].mode()[0]
print("Gender of most of the customers:", most_common_gender)

Gender of most of the customers: Female

24. What is the gender distribution per branch?

gender_distribution_per_branch = data.groupby(['Branch',
'Gender']).size()
print("Gender distribution per branch:")
print(gender_distribution_per_branch)

Gender distribution per branch:

Branch Gender
A Female 161
Male 179
B Female 162
Male 170
C Female 178
Male 150
dtype: int64

25. Which time of the day do customers give

most ratings?
most_rated_time_of_day = data.groupby('Time')['Rating'].sum().idxmax()
print("Time of the day when customers give the most ratings:",
most_rated_time_of_day)

Time of the day when customers give the most ratings: 19:48:00
26. Which time of the day do customers give
most ratings per branch?
most_rated_time_of_day_per_branch = data.groupby(['Branch', 'Time'])
['Rating'].sum().idxmax()
print("Time of the day when customers give the most ratings per
branch:", most_rated_time_of_day_per_branch)

Time of the day when customers give the most ratings per branch: ('C',
'10:23:00')

27. Which day of the week has the best avg

ratings?
best_avg_ratings_day_of_week = data.groupby(data['Date'].dt.dayofweek)
['Rating'].mean().idxmax()
print("Day of the week with the best average ratings:",
best_avg_ratings_day_of_week)

Day of the week with the best average ratings: 0

28. Which day of the week has the best average

ratings per branch?
best_avg_ratings_day_of_week_per_branch = data.groupby(['Branch',
data['Date'].dt.dayofweek])['Rating'].mean().idxmax()
print("Day of the week with the best average ratings per branch:",
best_avg_ratings_day_of_week_per_branch)

Day of the week with the best average ratings per branch: ('B', 0)

29. Are there any patterns or trends in sales

over time (by month, day of the week, or time of
day)?
# For example, we can visualize total sales over time
import matplotlib.pyplot as plt

# Extract month, day of the week, and hour from the date
data['Month'] = data['Date'].dt.month
data['DayOfWeek'] = data['Date'].dt.dayofweek
data['Hour'] = data['Time'].apply(lambda x: int(x.split(':')[0]))

# Total sales by month

total_sales_by_month = data.groupby('Month')['Total'].sum()

# Total sales by day of the week

total_sales_by_day_of_week = data.groupby('DayOfWeek')['Total'].sum()

# Total sales by hour of the day

total_sales_by_hour = data.groupby('Hour')['Total'].sum()

# Plotting
plt.figure(figsize=(18, 5))

plt.subplot(1, 3, 1)
plt.plot(total_sales_by_month, marker='o')
plt.title('Total Sales by Month')
plt.xlabel('Month')
plt.ylabel('Total Sales')

plt.subplot(1, 3, 2)
plt.plot(total_sales_by_day_of_week, marker='o')
plt.title('Total Sales by Day of the Week')
plt.xlabel('Day of the Week')
plt.ylabel('Total Sales')

plt.subplot(1, 3, 3)
plt.plot(total_sales_by_hour, marker='o')
plt.title('Total Sales by Hour of the Day')
plt.xlabel('Hour of the Day')
plt.ylabel('Total Sales')

plt.tight_layout()
plt.show()
30. Are there any differences in customer
ratings between branches?
ratings_by_branch = data.groupby('Branch')['Rating'].mean()
print("Average ratings by branch:")
print(ratings_by_branch)

Average ratings by branch:

Branch
A 7.027059
B 6.818072
C 7.072866
Name: Rating, dtype: float64

31. Is there any correlation between the tax

amount and the total transaction amount?
correlation_tax_total = data['Tax 5%'].corr(data['Total'])
print("Correlation between tax amount and total transaction amount:",
correlation_tax_total)

Correlation between tax amount and total transaction amount:

0.9999999999999998

32. Do certain product lines tend to have higher

ratings than others?
ratings_by_product_line = data.groupby('Product line')
['Rating'].mean()
print("Average ratings by product line:")
print(ratings_by_product_line)

Average ratings by product line:

Product line
Electronic accessories 6.924706
Fashion accessories 7.029213
Food and beverages 7.113218
Health and beauty 7.003289
Home and lifestyle 6.837500
Sports and travel 6.916265
Name: Rating, dtype: float64

33. Is there any correlation between the

quantity of items purchased and the total
transaction amount?
correlation_quantity_total = data['Quantity'].corr(data['Total'])
print("Correlation between quantity and total transaction amount:",
correlation_quantity_total)