Python Dictionaries

Recently, I was working on a data processing project where I needed to store and retrieve information quickly. The solution? Python dictionaries. These powerful data structures have been my go-to tool for organizing data for years.

In this guide, I’ll walk you through everything you need to know about Python dictionaries – from basics to advanced techniques. Whether you’re managing customer data, configuring settings, or analyzing sales information, dictionaries will make your code more efficient.

So let’s dive into Python dictionaries and unlock their full potential!

What is a Python Dictionary?

A Python dictionary is a collection of key-value pairs where each key must be unique. Think of it like a real-world dictionary where each word (key) has a definition (value).

Dictionaries are:

• Unordered (in Python versions before 3.7)
• Ordered (from Python 3.7 onwards)
• Mutable (can be changed)
• Indexed by keys, not positions

Check out all the tutorials related to the topic Python Arrays.

Creating a Python Dictionary

Method 1: Using Curly Braces {}

The most common way to create a dictionary is using curly braces:

# Empty dictionary
my_dict = {}

# Dictionary with initial values
customer = {
    'name': 'John Smith',
    'age': 35,
    'city': 'New York',
    'is_active': True
}

print(customer)

Method 2: Using the dict() Constructor

# Empty dictionary
my_dict = dict()

# Dictionary from sequence of key-value pairs
customer = dict([
    ('name', 'John Smith'),
    ('age', 35),
    ('city', 'New York'),
    ('is_active', True)
])

print(customer)

Method 3: Using Dictionary Comprehension

Dictionary comprehensions provide a concise way to create dictionaries:

# Create a dictionary of squares
squares = {x: x*x for x in range(1, 6)}
print(squares)  # Output: {1: 1, 2: 4, 3: 9, 4: 16, 5: 25}

# Create a dictionary from two lists
cities = ['New York', 'Los Angeles', 'Chicago']
populations = [8804190, 3898747, 2746388]
city_pops = {city: pop for city, pop in zip(cities, populations)}
print(city_pops)  # Output: {'New York': 8804190, 'Los Angeles': 3898747, 'Chicago': 2746388}

Accessing Dictionary Values

There are several ways to access values in a dictionary:

Using Square Brackets []

customer = {'name': 'John Smith', 'age': 35, 'city': 'New York'}

# Access a value
print(customer['name'])  # Output: John Smith

# This will raise KeyError if the key doesn't exist
# print(customer['email'])  # KeyError: 'email'

Using get() Method

The get() method is safer because it returns None (or a default value) if the key doesn’t exist:

customer = {'name': 'John Smith', 'age': 35, 'city': 'New York'}

# Access a value with get()
print(customer.get('name'))  # Output: John Smith

# Using get() with a non-existent key
print(customer.get('email'))  # Output: None

# Providing a default value
print(customer.get('email', 'Not provided'))  # Output: Not provided

Read more about the topic Python Conditional Statements and Loops in this tutorial.

Modifying Dictionaries

Dictionaries are mutable, which means we can change them after creation.

Adding or Updating Items

customer = {'name': 'John Smith', 'age': 35}

# Add a new key-value pair
customer['city'] = 'New York'

# Update an existing value
customer['age'] = 36

print(customer)  # Output: {'name': 'John Smith', 'age': 36, 'city': 'New York'}

Using update() Method

The update() method adds multiple key-value pairs at once:

customer = {'name': 'John Smith', 'age': 35}

# Update with multiple key-value pairs
customer.update({'city': 'New York', 'email': 'john@example.com', 'age': 36})

print(customer)  
# Output: {'name': 'John Smith', 'age': 36, 'city': 'New York', 'email': 'john@example.com'}

Removing Items

There are several ways to remove items from a dictionary:

customer = {'name': 'John Smith', 'age': 36, 'city': 'New York', 'email': 'john@example.com'}

# Remove a specific item and return its value
age = customer.pop('age')
print(age)  # Output: 36
print(customer)  # Output: {'name': 'John Smith', 'city': 'New York', 'email': 'john@example.com'}

# Remove and return the last inserted item (Python 3.7+)
last_item = customer.popitem()
print(last_item)  # Output: ('email', 'john@example.com')
print(customer)  # Output: {'name': 'John Smith', 'city': 'New York'}

# Delete a specific item
del customer['city']
print(customer)  # Output: {'name': 'John Smith'}

# Clear all items
customer.clear()
print(customer)  # Output: {}

Dictionary Methods

Python dictionaries come with several built-in methods that make them even more powerful:

keys(), values(), and items()

product = {
    'name': 'Laptop',
    'price': 999.99,
    'brand': 'TechPro',
    'in_stock': True
}

# Get all keys
print(list(product.keys()))  # Output: ['name', 'price', 'brand', 'in_stock']

# Get all values
print(list(product.values()))  # Output: ['Laptop', 999.99, 'TechPro', True]

# Get all key-value pairs as tuples
print(list(product.items()))  
# Output: [('name', 'Laptop'), ('price', 999.99), ('brand', 'TechPro'), ('in_stock', True)]

setdefault() Method

The setdefault() method returns the value of a key if it exists. If not, it inserts the key with a specified value and returns that value:

customer = {'name': 'John Smith', 'city': 'New York'}

# Get existing value
email = customer.setdefault('name', 'Unknown')
print(email)  # Output: John Smith

# Set default for non-existent key
email = customer.setdefault('email', 'no-email@example.com')
print(email)  # Output: no-email@example.com
print(customer)  
# Output: {'name': 'John Smith', 'city': 'New York', 'email': 'no-email@example.com'}

Learn more about the topic of Python Data Types.

Dictionary Comprehension with Conditionals

We can use conditionals in dictionary comprehensions for more advanced filtering:

prices = {'apple': 1.2, 'banana': 0.5, 'orange': 0.8, 'grapes': 2.5, 'watermelon': 5.0}

# Create dictionary of fruits that cost less than $1
affordable_fruits = {fruit: price for fruit, price in prices.items() if price < 1}
print(affordable_fruits)  # Output: {'banana': 0.5, 'orange': 0.8}

# Create dictionary with applied discount for expensive fruits
discounted_prices = {fruit: (price * 0.8 if price > 2 else price) 
                     for fruit, price in prices.items()}
print(discounted_prices)  
# Output: {'apple': 1.2, 'banana': 0.5, 'orange': 0.8, 'grapes': 2.0, 'watermelon': 4.0}

Nested Dictionaries

Dictionaries can contain other dictionaries as values, allowing for complex data structures:

employees = {
    'E001': {
        'name': 'John Smith',
        'department': 'Sales',
        'salary': 65000,
        'contact': {
            'email': 'john@example.com',
            'phone': '555-1234'
        }
    },
    'E002': {
        'name': 'Lisa Johnson',
        'department': 'Marketing',
        'salary': 70000,
        'contact': {
            'email': 'lisa@example.com',
            'phone': '555-5678'
        }
    }
}

# Accessing nested values
print(employees['E001']['name'])  # Output: John Smith
print(employees['E002']['contact']['email'])  # Output: lisa@example.com

# Modifying nested values
employees['E001']['salary'] = 68000
employees['E002']['contact']['phone'] = '555-9999'

print(employees['E001']['salary'])  # Output: 68000
print(employees['E002']['contact']['phone'])  # Output: 555-9999

Dictionary with Multiple Values per Key

Sometimes we need to store multiple values for a single key. There are several ways to achieve this:

Using Lists as Values

student_courses = {
    'John': ['Math', 'Physics', 'Computer Science'],
    'Lisa': ['History', 'English', 'Art'],
    'Mike': ['Biology', 'Chemistry']
}

# Adding a new course for John
student_courses['John'].append('Economics')
print(student_courses['John'])  # Output: ['Math', 'Physics', 'Computer Science', 'Economics']

# Adding a new student
student_courses['Sarah'] = ['Geography', 'Spanish']
print(student_courses)

Using Tuples as Values

If you don’t need to modify the values, tuples can be more efficient:

# Store multiple phone numbers
contacts = {
    'John': ('555-1234', '555-5678'),
    'Lisa': ('555-8765',),
    'Mike': ('555-4321', '555-8765', '555-9999')
}

# Access first phone number
print(contacts['John'][0])  # Output: 555-1234

Using Dictionaries as Values

For more structured data:

customers = {
    'C001': {
        'name': 'John Smith',
        'purchases': ['Laptop', 'Mouse', 'Keyboard'],
        'total_spent': 1250.75
    },
    'C002': {
        'name': 'Lisa Johnson',
        'purchases': ['Phone', 'Headphones'],
        'total_spent': 950.25
    }
}

# Add a new purchase for John
customers['C001']['purchases'].append('Monitor')
customers['C001']['total_spent'] += 299.99

print(customers['C001'])

Dictionary Counting and Frequency Analysis

Dictionaries are excellent for counting occurrences:

# Count word frequency in a text
text = "how much wood would a woodchuck chuck if a woodchuck could chuck wood"
words = text.split()

word_count = {}
for word in words:
    word_count[word] = word_count.get(word, 0) + 1

print(word_count)
# Output: {'how': 1, 'much': 1, 'wood': 2, 'would': 1, 'a': 2, 'woodchuck': 2, 'chuck': 2, 'if': 1, 'could': 1}

This technique, known as frequency analysis, is incredibly useful for data processing tasks.

Finding the Most Common Items

We can use dictionaries to find the most common items:

# Find the most common word
most_common_word = max(word_count.items(), key=lambda x: x[1])
print(f"Most common word: {most_common_word[0]} (appears {most_common_word[1]} times)")
# Output: Most common word: wood (appears 2 times) - or any other word with count 2

Using Counter from Collections

The collections module provides a specialized Counter class that makes frequency counting even easier:

from collections import Counter

# Same text as before
text = "how much wood would a woodchuck chuck if a woodchuck could chuck wood"
words = text.split()

# Count word frequency using Counter
word_count = Counter(words)
print(word_count)
# Output: Counter({'wood': 2, 'a': 2, 'woodchuck': 2, 'chuck': 2, 'how': 1, 'much': 1, 'would': 1, 'if': 1, 'could': 1})

# Find most common words
most_common = word_count.most_common(3)
print(most_common)  # Output: [('wood', 2), ('a', 2), ('woodchuck', 2)] - order may vary

Dictionary Performance and Memory Usage

One of the reasons I love dictionaries is their performance. Looking up values by key is extremely fast (O(1) on average), making them perfect for large datasets.

Here’s an example where I compare list search vs. dictionary lookup for finding sales data:

import time

# Create test data - 10,000 products
product_ids = list(range(10000))
sales_amounts = [i * 10 for i in range(10000)]

# Create list of tuples (slower to search)
sales_list = list(zip(product_ids, sales_amounts))

# Create dictionary (faster to search)
sales_dict = dict(zip(product_ids, sales_amounts))

# Search in list - O(n) time complexity
start_time = time.time()
for _ in range(1000):
    product_id = 9876  # Product near the end of the list
    for item in sales_list:
        if item[0] == product_id:
            sale_amount = item[1]
            break
list_time = time.time() - start_time

# Search in dictionary - O(1) time complexity
start_time = time.time()
for _ in range(1000):
    product_id = 9876
    sale_amount = sales_dict[product_id]
dict_time = time.time() - start_time

print(f"List search time: {list_time:.6f} seconds")
print(f"Dictionary lookup time: {dict_time:.6f} seconds")
print(f"Dictionary is {list_time/dict_time:.1f}x faster")

Practical Applications of Dictionaries

Over my years of Python development, I’ve used dictionaries in countless applications. Here are some real-world examples:

1. Configuration Settings

app_config = {
    'database': {
        'host': 'localhost',
        'port': 5432,
        'username': 'admin',
        'password': 'secure_password',
        'database_name': 'product_db'
    },
    'api': {
        'base_url': 'https://api.example.com',
        'version': 'v2',
        'timeout': 30,
        'retry_limit': 3
    },
    'logging': {
        'level': 'INFO',
        'file_path': '/var/log/app.log',
        'rotation': '1 day'
    }
}

# Access configuration values
db_host = app_config['database']['host']
api_timeout = app_config['api']['timeout']

2. Data Processing and Transformation

# Process customer orders
orders = [
    {'customer_id': 'C001', 'product': 'Laptop', 'amount': 999.99},
    {'customer_id': 'C002', 'product': 'Phone', 'amount': 699.99},
    {'customer_id': 'C001', 'product': 'Headphones', 'amount': 149.99},
    {'customer_id': 'C003', 'product': 'Tablet', 'amount': 349.99},
    {'customer_id': 'C002', 'product': 'Case', 'amount': 29.99}
]

# Calculate total spending per customer
customer_totals = {}
for order in orders:
    customer_id = order['customer_id']
    amount = order['amount']
    customer_totals[customer_id] = customer_totals.get(customer_id, 0) + amount

print(customer_totals)
# Output: {'C001': 1149.98, 'C002': 729.98, 'C003': 349.99}

3. Caching and Memoization

# Using a dictionary as a cache for expensive function calls
fibonacci_cache = {}

def fibonacci(n):
    # Check if value exists in cache
    if n in fibonacci_cache:
        return fibonacci_cache[n]
    
    # Calculate the nth term
    if n <= 1:
        value = n
    else:
        value = fibonacci(n-1) + fibonacci(n-2)
    
    # Cache the result and return
    fibonacci_cache[n] = value
    return value

# Calculate first 20 Fibonacci numbers
for i in range(20):
    print(f"Fibonacci({i}) = {fibonacci(i)}")

4. Implementing a Simple Database

# Simple in-memory database using dictionaries
users_db = {}

def add_user(user_id, name, email, age):
    users_db[user_id] = {
        'name': name,
        'email': email,
        'age': age,
        'created_at': time.time()
    }
    return f"User {name} added successfully"

def get_user(user_id):
    return users_db.get(user_id, "User not found")

def update_user(user_id, **kwargs):
    if user_id in users_db:
        users_db[user_id].update(kwargs)
        return f"User {user_id} updated successfully"
    return "User not found"

def delete_user(user_id):
    if user_id in users_db:
        del users_db[user_id]
        return f"User {user_id} deleted successfully"
    return "User not found"

# Usage example
add_user("U001", "John Smith", "john@example.com", 35)
add_user("U002", "Lisa Johnson", "lisa@example.com", 28)
print(get_user("U001"))
update_user("U001", email="john.smith@example.com", age=36)
print(get_user("U001"))

Read all the tutorials related to the topic of Python Functions.

Counting Frequencies Using Dictionaries

Counting the frequencies of items in collections is one of the most common use cases for dictionaries. Let’s explore this further:

# Analyzing US state populations (sample data)
states = ['California', 'Texas', 'Florida', 'New York', 'Texas', 'Florida', 
          'California', 'Texas', 'Georgia', 'California', 'Washington']

# Count occurrences using a dictionary
state_counts = {}
for state in states:
    if state in state_counts:
        state_counts[state] += 1
    else:
        state_counts[state] = 1

print(state_counts)
# Output: {'California': 3, 'Texas': 3, 'Florida': 2, 'New York': 1, 'Georgia': 1, 'Washington': 1}

# More elegantly with get() method
state_counts = {}
for state in states:
    state_counts[state] = state_counts.get(state, 0) + 1

print(state_counts)

Working with JSON and Dictionaries

Python dictionaries are closely related to JSON, making them perfect for web applications:

import json

# Dictionary representing data for a weather API
weather_data = {
    'location': {
        'city': 'New York',
        'state': 'NY',
        'country': 'USA',
        'coordinates': {
            'latitude': 40.7128,
            'longitude': -74.0060
        }
    },
    'current': {
        'temperature': 72.5,
        'feels_like': 74.2,
        'humidity': 65,
        'wind_speed': 10.4,
        'conditions': 'Partly Cloudy'
    },
    'forecast': [
        {'day': 'Monday', 'high': 75, 'low': 65, 'conditions': 'Sunny'},
        {'day': 'Tuesday', 'high': 80, 'low': 68, 'conditions': 'Clear'},
        {'day': 'Wednesday', 'high': 77, 'low': 66, 'conditions': 'Partly Cloudy'}
    ]
}

# Convert dictionary to JSON string
json_data = json.dumps(weather_data, indent=4)
print(json_data)

# Convert JSON string back to dictionary
parsed_data = json.loads(json_data)
print(parsed_data['location']['city'])  # Output: New York
print(parsed_data['forecast'][1]['high'])  # Output: 80

Advanced Dictionary Techniques

Dictionary Union (Python 3.9+)

In Python 3.9 and later, you can use the | operator to merge dictionaries:

# Base product information
product = {
    'id': 'P001',
    'name': 'Smartphone',
    'price': 799.99
}

# Additional product details
details = {
    'brand': 'TechX',
    'model': 'X20',
    'color': 'Midnight Blue'
}

# Merge dictionaries with |
complete_product = product | details
print(complete_product)
# Output: {'id': 'P001', 'name': 'Smartphone', 'price': 799.99, 'brand': 'TechX', 'model': 'X20', 'color': 'Midnight Blue'}

# Update and merge with |=
product |= details
print(product)
# Output: {'id': 'P001', 'name': 'Smartphone', 'price': 799.99, 'brand': 'TechX', 'model': 'X20', 'color': 'Midnight Blue'}

Using defaultdict

The defaultdict from the collections module creates dictionaries with default values for missing keys:

from collections import defaultdict

# Using defaultdict with int as default factory
word_count = defaultdict(int)
for word in "how much wood would a woodchuck chuck".split():
    word_count[word] += 1  # No need to check if key exists

print(dict(word_count))
# Output: {'how': 1, 'much': 1, 'wood': 1, 'would': 1, 'a': 1, 'woodchuck': 1, 'chuck': 1}

The default dict in Python behaves almost exactly like a regular dictionary, but automatically assigns default values when accessing missing keys. This makes certain patterns much cleaner.

Using defaultdict with list

Another common use case is creating a dictionary where each value is a list:

# Group customers by state
customers = [
    ('John Smith', 'NY'),
    ('Lisa Johnson', 'CA'),
    ('Michael Brown', 'TX'),
    ('Sarah Davis', 'CA'),
    ('Robert Wilson', 'NY'),
    ('Jennifer Lee', 'TX'),
    ('David Garcia', 'FL')
]

# Without defaultdict (traditional approach)
customers_by_state = {}
for name, state in customers:
    if state not in customers_by_state:
        customers_by_state[state] = []
    customers_by_state[state].append(name)

# With defaultdict (cleaner approach)
customers_by_state = defaultdict(list)
for name, state in customers:
    customers_by_state[state].append(name)

print(dict(customers_by_state))
# Output: {'NY': ['John Smith', 'Robert Wilson'], 'CA': ['Lisa Johnson', 'Sarah Davis'], 'TX': ['Michael Brown', 'Jennifer Lee'], 'FL': ['David Garcia']}

Nested defaultdicts

You can even create nested defaultdicts for more complex data structures:

# Sales data by region, quarter, and product
sales = defaultdict(lambda: defaultdict(lambda: defaultdict(int)))

# Adding sales data
sales['West']['Q1']['Laptop'] += 50
sales['West']['Q1']['Phone'] += 100
sales['East']['Q1']['Laptop'] += 30
sales['West']['Q2']['Laptop'] += 45
sales['East']['Q2']['Phone'] += 75

# Accessing data (no KeyError even for new combinations)
print(sales['West']['Q1']['Laptop'])  # Output: 50
print(sales['South']['Q3']['Tablet'])  # Output: 0 (default value)

# Convert to regular dict for display
print(dict(sales))

Common Dictionary Patterns

Dictionary Sorting

Dictionaries in Python 3.7+ maintain insertion order, but sometimes we need to sort them:

# Sort dictionary by keys
product_prices = {
    'Laptop': 999.99,
    'Phone': 699.99,
    'Headphones': 149.99,
    'Tablet': 349.99,
    'Smartwatch': 249.99
}

# Sort by keys (alphabetical)
sorted_by_name = dict(sorted(product_prices.items()))
print(sorted_by_name)

# Sort by values (price)
sorted_by_price = dict(sorted(product_prices.items(), key=lambda item: item[1]))
print(sorted_by_price)

# Sort by values descending (highest price first)
sorted_by_price_desc = dict(sorted(product_prices.items(), 
                                  key=lambda item: item[1], 
                                  reverse=True))
print(sorted_by_price_desc)

Dictionary Filtering

Often, we need to filter dictionaries based on certain conditions:

sales_data = {
    'Laptop': 50,
    'Phone': 100,
    'Tablet': 30,
    'Desktop': 15,
    'Headphones': 150,
    'Keyboard': 75
}

# Filter items with sales >= 50 (traditional way)
high_volume_items = {}
for product, count in sales_data.items():
    if count >= 50:
        high_volume_items[product] = count

print(high_volume_items)
# Output: {'Laptop': 50, 'Phone': 100, 'Headphones': 150, 'Keyboard': 75}

# Dictionary comprehension way (more concise)
high_volume_items = {product: count for product, count in sales_data.items() if count >= 50}
print(high_volume_items)

Dictionary Transformation

Transforming dictionary values is another common operation:

prices = {
    'Laptop': 999.99,
    'Phone': 699.99,
    'Headphones': 149.99,
    'Tablet': 349.99
}

# Apply 10% discount to all products
discounted_prices = {product: price * 0.9 for product, price in prices.items()}
print(discounted_prices)

# Format prices as strings with $ symbol
formatted_prices = {product: f"${price:.2f}" for product, price in prices.items()}
print(formatted_prices)
# Output: {'Laptop': '$999.99', 'Phone': '$699.99', 'Headphones': '$149.99', 'Tablet': '$349.99'}

Check out the Python Operators to learn more about operators.

Dictionary Views and Iterating

Dictionary views (keys(), values(), and items()) provide dynamic views of dictionary contents:

inventory = {
    'Laptop': 25,
    'Phone': 50,
    'Tablet': 30
}

# Get views
keys_view = inventory.keys()
values_view = inventory.values()
items_view = inventory.items()

print(keys_view)    # dict_keys(['Laptop', 'Phone', 'Tablet'])
print(values_view)  # dict_values([25, 50, 30])
print(items_view)   # dict_items([('Laptop', 25), ('Phone', 50), ('Tablet', 30)])

# Views are dynamic - they reflect dictionary changes
inventory['Headphones'] = 45
print(keys_view)    # dict_keys(['Laptop', 'Phone', 'Tablet', 'Headphones'])

# Common dictionary iteration patterns
# Iterating through keys (default behavior)
for product in inventory:
    print(product)

# Iterating through values
for quantity in inventory.values():
    print(quantity)

# Iterating through key-value pairs
for product, quantity in inventory.items():
    print(f"{product}: {quantity} units")

Dictionary Tricks for Data Analysis

As someone who frequently deals with data analysis, these dictionary tricks have proved invaluable:

Grouping and Aggregating Data

# Sales transactions
transactions = [
    {"product": "Laptop", "price": 999.99, "region": "West"},
    {"product": "Phone", "price": 699.99, "region": "East"},
    {"product": "Laptop", "price": 1199.99, "region": "East"},
    {"product": "Headphones", "price": 149.99, "region": "West"},
    {"product": "Phone", "price": 599.99, "region": "West"},
    {"product": "Tablet", "price": 349.99, "region": "East"},
    {"product": "Laptop", "price": 899.99, "region": "West"}
]

# Group total sales by region
sales_by_region = {}
for t in transactions:
    region = t["region"]
    if region not in sales_by_region:
        sales_by_region[region] = 0
    sales_by_region[region] += t["price"]

print(sales_by_region)
# Output: {'West': 2649.96, 'East': 2249.97}

# Group sales by product type
sales_by_product = {}
for t in transactions:
    product = t["product"]
    if product not in sales_by_product:
        sales_by_product[product] = []
    sales_by_product[product].append(t["price"])

# Calculate average price per product
avg_price_by_product = {product: sum(prices)/len(prices) 
                        for product, prices in sales_by_product.items()}

print(avg_price_by_product)
# Output: {'Laptop': 1033.32, 'Phone': 649.99, 'Headphones': 149.99, 'Tablet': 349.99}

Dictionary Performance Optimization

Using dict.fromkeys()

When you need to create a dictionary with the same value for multiple keys:

# Initialize inventory with zero quantities (less efficient)
products = ["Laptop", "Phone", "Tablet", "Headphones", "Monitor", "Keyboard", "Mouse"]
inventory = {}
for product in products:
    inventory[product] = 0

# Better way using dict.fromkeys()
inventory = di