Complete Pandas Tutorial in PythonMind map · Lesson 13 of 13

Learn Data Cleaning in Pandas – Mind Map

Everything this lesson covers — Audit Data, Fix Data, Code Demonstration (Sample 1) and Code Demonstration (Sample 2) — in one visual map. Click any topic to open it, or download the full map as an image for revision.

Watch this lesson in the course Watch on YouTube

Mind map

The Data Cleaning Way

Click a topic to open or close it · drag to move · scroll to zoom · the download includes every topic, fully expanded

Learn Data Cleaning in Pandas: lesson notes

01Audit Data
- Read Data
  - df.shape
  - df.columns
  - df.info()
- Check Missing Values
  - df.isnull()
  - df.isnull().sum()
  - df.isnull().sum().sum()
- Check Duplicate Rows
  - df.duplicated()
  - df.duplicated().sum()
  - df.duplicated(subset=['column_name'], keep='first/last/false')
- Check Outliers
  - Identify inconsistent data
  - Filter data based on conditions (e.g., df[df['age'] < 100])
- Check Data Types
  - df.info() for dtypes
  - df.memory_usage(deep=True) for memory consumption
02Fix Data
- Fill Missing Values
  - df['column'].fillna(value)
  - Use mean/median for numerical data
  - Use mode for categorical data
  - Use 'Unknown' or other placeholder for strings
- Remove Duplicate Values

df.drop_duplicates()

df.drop_duplicates(subset=['column_name'])

Fix Outliers

Filter out extreme values
Replace with mean/median/mode

Handle Inconsistencies

df['column'].str.strip() to remove spaces
df['column'].str.title() for consistent casing
Convert data types (e.g., to_datetime)

03Code Demonstration (Sample 1)

Import Libraries
- import pandas as pd
- import numpy as np
Create DataFrame 'df'
- pd.DataFrame(data)
Audit & Fix Steps
- df.shape, df.columns, df.info()
- df.isnull().sum()
- df['salary'].fillna(df['salary'].mean())
- df['name'].fillna('Unknown')
- df.drop_duplicates()
- df['name'].str.strip().str.title()
- df[df['age'] < 100]

04Code Demonstration (Sample 2)

Create DataFrame 'raw_orders'
- pd.DataFrame(data)
- raw_orders.copy() to 'original'
Audit Steps
- raw_orders.shape
- raw_orders.info()
- raw_orders.memory_usage(deep=True).sum()
- raw_orders.isnull().sum()
- Calculate missing percentage
- raw_orders.duplicated().sum()
- raw_orders.duplicated(subset=['order_id'], keep='first/last/false')
- raw_orders['product_category'].value_counts(dropna=False)
- raw_orders[['quantity', 'unit_price', 'total_amount']].describe().round(1)
- raw_orders[raw_orders['quantity'] < 0]
- Check consistency: expected_total = quantity * unit_price
Fix Steps
- orders = raw_orders.copy()
- orders['quantity'].fillna(orders['quantity'].median())
- orders['product_category'].fillna(orders['product_category'].mode()[0])
- orders['status'].fillna('Unknown')
- orders.dropna(subset=['customer_name', 'city'])

Learn Data Cleaning in Pandas – Mind Map

The Data Cleaning Way

Learn Data Cleaning in Pandas: lesson notes

01Audit Data

Read Data

Check Missing Values

Check Duplicate Rows

Check Outliers

Check Data Types

02Fix Data

Fill Missing Values

Remove Duplicate Values

Fix Outliers

Handle Inconsistencies

03Code Demonstration (Sample 1)

Import Libraries

Create DataFrame 'df'

Audit & Fix Steps

04Code Demonstration (Sample 2)

Create DataFrame 'raw_orders'

Audit Steps

Fix Steps