5/8/25, 11:51 AM Rough work.

ipynb - Colab

1) Data Exploration: Load the dataset and display the first few rows. What are the columns present in the dataset? Check for null values and
provide a summary of the dataset.

import pandas as pd

# Load the dataset

df = pd.read_csv('/content/drive/MyDrive/[Link]')

# Display the first few rows of the dataset

print("First few rows of the dataset:")
print([Link]())

# List the columns present in the dataset

print("\nColumns in the dataset:")
print([Link])

# Check for missing values in each column

print("\nMissing values in each column:")
print([Link]().sum())

# Provide summary statistics of the dataset

print("\nSummary statistics of the dataset:")
print([Link]())

First few rows of the dataset:

TransactionNo Items DateTime Daypart DayType Quantity \
0 1 Bread 30-10-2016 09:58 Morning Weekend 5
1 2 Scandinavian 30-10-2016 10:05 Morning Weekend 1
2 2 Scandinavian 30-10-2016 10:05 Morning Weekend 1
3 3 Hot chocolate 30-10-2016 10:07 Morning Weekend 4
4 3 Jam 30-10-2016 10:07 Morning Weekend 3

Price
0 35
1 43
2 43
3 97
4 96

Columns in the dataset:

Index(['TransactionNo', 'Items', 'DateTime', 'Daypart', 'DayType', 'Quantity',
'Price'],
dtype='object')

Missing values in each column:

TransactionNo 0
Items 0
DateTime 0
Daypart 0
DayType 0
Quantity 0
Price 0
dtype: int64

Summary statistics of the dataset:

TransactionNo Quantity Price
count 20507.000000 20507.000000 20507.000000
mean 4976.202370 2.998147 51.027454
std 2796.203001 1.415246 17.976061
min 1.000000 1.000000 31.000000
25% 2552.000000 2.000000 39.000000
50% 5137.000000 3.000000 43.000000
75% 7357.000000 4.000000 57.000000
max 9684.000000 5.000000 99.000000

2) Data Cleaning: Are there any missing or duplicate values in the dataset? If so, clean the data accordingly.

# Display the first few rows to understand the structure

print("First few rows of the dataset:")
print([Link]())

# Check for missing values

missing_values = [Link]().sum()

# Print columns with missing values

print("\nMissing values in each column:")
print(missing_values)

# Handle missing values:

[Link] 1/6
5/8/25, 11:51 AM Rough [Link] - Colab
# For numerical columns (Quantity, Price), we can replace missing values with the median.
df['Quantity'].fillna(df['Quantity'].median(), inplace=True)
df['Price'].fillna(df['Price'].median(), inplace=True)

# For categorical columns (Items, Daypart, DayType), we can replace missing values with the mode (most frequent value).
df['Items'].fillna(df['Items'].mode()[0], inplace=True)
df['Daypart'].fillna(df['Daypart'].mode()[0], inplace=True)
df['DayType'].fillna(df['DayType'].mode()[0], inplace=True)

# # For DateTime column, if it's missing, it can be replaced with a placeholder date (or dropped, depending on use case).
# df['DateTime'].fillna('Unknown', inplace=True)

# Check for duplicate rows

duplicate_rows = [Link]().sum()

# Print the number of duplicate rows

print("\nNumber of duplicate rows:", duplicate_rows)

# Remove duplicate rows if any

df.drop_duplicates(inplace=True)

# Verify that missing values and duplicates are handled

print("\nAfter cleaning:")
print([Link]().sum()) # Check for remaining missing values
print("\nNumber of duplicate rows after cleaning:", [Link]().sum())

# Display the first few rows again after cleaning

print("\nFirst few rows after cleaning:")
print([Link]())

Items 0 
DateTime 0
Daypart 0
DayType 0
Quantity 0
Price 0
dtype: int64

Number of duplicate rows: 0

After cleaning:
TransactionNo 0
Items 0
DateTime 0
Daypart 0
DayType 0
Quantity 0
Price 0
dtype: int64

Number of duplicate rows after cleaning: 0

First few rows after cleaning:

TransactionNo Items DateTime Daypart DayType Quantity \
0 1 Bread 30-10-2016 09:58 Morning Weekend 5
1 2 Scandinavian 30-10-2016 10:05 Morning Weekend 1
3 3 Hot chocolate 30-10-2016 10:07 Morning Weekend 4
4 3 Jam 30-10-2016 10:07 Morning Weekend 3
5 3 Cookies 30-10-2016 10:07 Morning Weekend 3

Price
0 35
1 43
3 97
4 96
5 72
<ipython-input-26-38dd57f669c1>:14: FutureWarning: A value is trying to be set on a copy of a DataFrame or Series through chained
The behavior will change in pandas 3.0. This inplace method will never work because the intermediate object on which we are setti

df['Items'] fillna(df['Items'] mode()[0] inplace=True) 

[Link] 2/6
5/8/25, 11:51 AM Rough [Link] - Colab
df[ Items ].fillna(df[ Items ].mode()[0], inplace=True)


3) Sales Analysis: Which product is the most sold item in the bakery? Show the top 5 products based on sales volume. (hint sort the datframe
in descending order, store it in a new dataframe and then use head function on it)

# Group the data by 'Items' (product) and calculate the total sales volume (sum of Quantity)
sales_volume = [Link]('Items')['Quantity'].sum().reset_index()

# Sort the products by sales volume in descending order

sorted_sales = sales_volume.sort_values(by='Quantity', ascending=False)

# Get the top 5 most sold products

top_5_products = sorted_sales.head(5)

# Display the top 5 products

print("Top 5 products based on sales volume:")
print(top_5_products)

Top 5 products based on sales volume:

Items Quantity
23 Coffee 15763
11 Bread 9811
83 Tea 4162
15 Cake 3023
65 Pastry 2592

4) Revenue Generation: Calculate the total revenue generated by each product. Which product contributed the most to the bakery’s revenue?

# Calculate revenue for each transaction (Quantity * Price)

df['Revenue'] = df['Quantity'] * df['Price']

# Group the data by 'Items' and calculate total revenue for each product
revenue_by_product = [Link]('Items')['Revenue'].sum().reset_index()

# Sort the products by total revenue in descending order

sorted_revenue = revenue_by_product.sort_values(by='Revenue', ascending=False)

# Get the product that contributed the most to the bakery’s revenue
top_revenue_product = sorted_revenue.head(1)

# Display the product with the highest revenue

print("Product that contributed the most to the bakery’s revenue:")
print(top_revenue_product)

# Optionally, display the top 5 products based on revenue

top_5_revenue_products = sorted_revenue.head(5) 
print("\nTop
 5 products based on revenue:") 
print(top_5_revenue_products)

Product that contributed the most to the bakery’s revenue:

Items Revenue
23 Coffee 677809

Top 5 products based on revenue:

Items Revenue
23 Coffee 677809
11 Bread 343385
48 Hot chocolate 171884
83 Tea 162318
15 Cake 148127

5) Time-based Analysis: Analyze sales trends over time. Which day of the week had the highest sales volume? Provide a visualization using
Matplotlib.

import [Link] as plt

# Convert the 'DateTime' column to datetime format

df['DateTime'] = pd.to_datetime(df['DateTime'])

# Extract the day of the week from the 'DateTime' column (0 = Monday, 6 = Sunday)
df['DayOfWeek'] = df['DateTime'].[Link]

# Group the data by 'DayOfWeek' and calculate total sales volume (sum of 'Quantity') for each day
l b d df b (' f k')[' i '] () i d ()
[Link] 3/6
5/8/25, 11:51 AM Rough [Link] - Colab
sales_by_day = [Link]('DayOfWeek')['Quantity'].sum().reset_index()

# Map numeric days to actual day names

sales_by_day['DayName'] = sales_by_day['DayOfWeek'].map({0: 'Monday', 1: 'Tuesday', 2: 'Wednesday', 3: 'Thursday',
4: 'Friday', 5: 'Saturday', 6: 'Sunday'})

# Sort the data by sales volume in descending order

sales_by_day_sorted = sales_by_day.sort_values(by='Quantity', ascending=False)

# Plot the sales volume for each day of the week

[Link](figsize=(10, 6))
[Link](sales_by_day_sorted['DayName'], sales_by_day_sorted['Quantity'], color='skyblue')

# Adding titles and labels

[Link]('Sales Volume by Day of the Week', fontsize=14)
[Link]('Day of the Week', fontsize=12)
[Link]('Total Sales Volume', fontsize=12)
[Link](rotation=45)
plt.tight_layout()

# Show the plot

[Link]()

# Display the day with the highest sales volume

print("Day with the highest sales volume:")
print(sales_by_day_sorted.head(1))

<ipython-input-29-f404eaa81e39>:5: UserWarning: Parsing dates in %d-%m-%Y %H:%M format when dayfirst=False (the default) was specifi
df['DateTime'] = pd.to_datetime(df['DateTime'])

Day with the highest sales volume:

DayOfWeek Quantity DayName
5 5 10453 Saturday

6) Product Sales Visualization: Create a bar chart using Seaborn to visualize the sales of the top 10 products.

import seaborn as sns

import [Link] as plt

# Group the data by 'Items' and calculate total sales volume (sum of 'Quantity') for each product
sales_by_product = [Link]('Items')['Quantity'].sum().reset_index()

# Sort the products by sales volume in descending order and get the top 10 products
top_10_products = sales_by_product.sort_values(by='Quantity', ascending=False).head(10)

# Create a Seaborn bar plot to visualize the sales of the top 10 products

[Link] 4/6
5/8/25, 11:51 AM Rough [Link] - Colab
[Link](figsize=(10, 6))
[Link](x='Quantity', y='Items', data=top_10_products, palette='viridis')

# Adding titles and labels

[Link]('Top 10 Products by Sales Volume', fontsize=14)
[Link]('Sales Volume (Quantity)', fontsize=12)
[Link]('Product', fontsize=12)
plt.tight_layout()

# Show the plot

[Link]()

<ipython-input-30-b0e4bf0ffe39>:14: FutureWarning:

Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `y` variable to `hue` and set `le

[Link](x='Quantity', y='Items', data=top_10_products, palette='viridis')

 

7) Correlation Analysis: If the dataset contains numerical features, analyze the correlation between them. Create a heatmap using Seaborn to
show the correlations.

# Select numerical columns for correlation analysis

numerical_columns = df.select_dtypes(include=['float64', 'int64']).columns

# Calculate the correlation matrix

correlation_matrix = df[numerical_columns].corr()

# Create a heatmap to visualize the correlations

[Link](figsize=(10, 8))
[Link](correlation_matrix, annot=True, cmap='coolwarm', fmt='.2f', linewidths=0.5, vmin=-1, vmax=1)

# Adding title and labels

[Link]('Correlation Heatmap', fontsize=14)
plt.tight_layout()

# Show the plot

[Link]()

[Link] 5/6
5/8/25, 11:51 AM Rough [Link] - Colab

Start coding or generate with AI.

[Link] 6/6