Chapter 1: Getting Started

Installation, IDEs, and Your First Script

1.1 Why Python for Analytics?

Python has become the default language for data science in industry and academia. It is open source, has a massive ecosystem of analytics libraries, and reads almost like pseudocode. Whether you plan to work in consulting, supply chain, finance, or tech, Python literacy gives you a significant advantage.

1.2 Installing Python

The two most common installation routes are Anaconda (recommended for beginners) and pip with python.org. Anaconda bundles Python with 250+ data science packages and the conda package manager.

Option A: Anaconda (Recommended)

Download the installer from anaconda.com/download. After installation, open the Anaconda Navigator to launch Jupyter Notebook or Spyder.

# Verify installation in your terminal
conda --version
python --version

Option B: pip + python.org

Download from python.org/downloads. After installing, use pip to add packages as needed.

# Install common data science packages
pip install numpy pandas matplotlib seaborn scikit-learn statsmodels jupyter

1.3 Virtual Environments

A virtual environment is an isolated Python installation that keeps each project's packages separate. Without environments, installing a package for one project can break another project that depends on a different version. Getting into the habit of creating one environment per project prevents "dependency hell" as your projects grow.

Using venv (Built-in)

Python ships with venv in the standard library. It creates a lightweight directory containing a private copy of the Python interpreter and its own pip.

# Create a virtual environment called 'myenv'
python -m venv myenv

# Activate it
# macOS / Linux:
source myenv/bin/activate

# Windows:
myenv\Scripts\activate

# Your prompt changes to show the active env
(myenv) $ pip install pandas numpy

# Deactivate when done
deactivate

Using conda Environments

If you installed Anaconda or Miniconda, conda manages environments and can install non-Python dependencies (C libraries, R, etc.) that pip cannot handle.

# Create a conda environment with a specific Python version
conda create -n analytics python=3.11

# Activate it
conda activate analytics

# Install packages (conda can mix conda + pip)
conda install pandas numpy scikit-learn
pip install statsmodels

# List all environments
conda env list

# Export environment for reproducibility
conda env export > environment.yml

# Recreate from file on another machine
conda env create -f environment.yml

pip vs conda: Quick Comparison

1.4 Choosing an IDE

Jupyter Magic Commands

Jupyter notebooks support "magic" commands (prefixed with % or !) that are not regular Python. These are useful for timing code, installing packages, or configuring plots without leaving the notebook.

# Time a single statement (runs it many times and reports average)
%timeit sum(range(10000))

# Time an entire cell (put %%timeit at the top of the cell)
%%timeit
total = 0
for i in range(10000):
    total += i

# Display matplotlib plots inline (run once at top of notebook)
%matplotlib inline

# Install a package directly from a notebook cell
!pip install seaborn

# Run a shell command
!ls *.csv

# Show all variables in the current namespace
%whos

# Reset all variables (start fresh)
%reset -f

1.5 Your First Script

Create a new file called hello.py or open a Jupyter cell and type the following:

# hello.py — Your first Python script
print("Hello, Data Science!")

# Variables and basic arithmetic
revenue = 150000
cost    = 95000
profit  = revenue - cost

print(f"Revenue: ${revenue:,}")
print(f"Cost:    ${cost:,}")
print(f"Profit:  ${profit:,}")

1.6 Variables and print()

Python variables do not need type declarations. The interpreter infers the type from the assigned value. Use print() to display output and f-strings for formatting.

# Variable assignment — no type declaration needed
product_name = "Widget A"       # str
units_sold   = 1250              # int
unit_price   = 29.99             # float
in_stock     = True              # bool

# f-string formatting
total = units_sold * unit_price
print(f"{product_name}: sold {units_sold} units at ${unit_price} each")
print(f"Total revenue: ${total:,.2f}")

# Check type
print(type(unit_price))  # <class 'float'>

1.7 type() and Type Checking

Since Python is dynamically typed, variables can change type at any time. The type() function tells you what type a value currently holds, and isinstance() checks whether a value belongs to a specific type. Understanding types helps you debug unexpected behavior, especially when reading data from files where numbers might arrive as strings.

quantity = 42
print(type(quantity))           # <class 'int'>

quantity = "forty-two"
print(type(quantity))           # <class 'str'>  — type changed!

# isinstance() is preferred for type checking
price = 19.99
print(isinstance(price, float))      # True
print(isinstance(price, (int, float))) # True (check multiple types)

# Type conversion (casting)
user_input = "150"
units = int(user_input)        # Convert string to int
rate  = float("3.14")          # Convert string to float
label = str(2024)              # Convert int to string

1.8 User Input with input()

The input() function pauses the program and waits for the user to type something. It always returns a string, so you need to convert the result if you expect a number.

# input() always returns a string
name = input("Enter product name: ")
qty  = int(input("Enter quantity: "))
cost = float(input("Enter unit cost: "))

total = qty * cost
print(f"Order for {qty} x {name}: ${total:,.2f}")

1.9 Comments and Code Style

Good code is readable code. Follow PEP 8 conventions: use snake_case for variables, add comments to explain why (not what), and keep lines under 79 characters.

# GOOD: descriptive variable names
order_quantity = 500
lead_time_days = 14

# AVOID: cryptic single-letter names
# q = 500
# l = 14

1.10 Common Beginner Errors

Every beginner encounters the same handful of errors. Recognizing them quickly will save you hours of frustration. Read the error message carefully, starting from the bottom line, which tells you the type of error and a brief description.

# 1. IndentationError — Python uses whitespace to define code blocks
if True:
print("oops")  # IndentationError: expected an indented block

# Fix: indent the body of if/for/while/def with 4 spaces
if True:
    print("fixed")

# 2. NameError — using a variable before defining it
print(total_cost)  # NameError: name 'total_cost' is not defined

# 3. TypeError — operating on incompatible types
result = "Price: " + 29.99  # TypeError: can only concatenate str to str
result = "Price: " + str(29.99)  # Fix: convert to string first

# 4. SyntaxError — forgetting colons, quotes, or parentheses
if x > 5     # SyntaxError: expected ':'
print("hello"  # SyntaxError: unexpected EOF (missing closing paren)

# 5. IndexError — accessing a list position that doesn't exist
items = ["A", "B", "C"]
print(items[5])  # IndexError: list index out of range

# 6. ModuleNotFoundError — package not installed
import pandas  # ModuleNotFoundError: No module named 'pandas'
# Fix: pip install pandas  (in your terminal, not in Python)

# 7. Mutable default argument trap (subtle!)
def add_item(item, basket=[]):  # BAD: default list is shared
    basket.append(item)
    return basket

print(add_item("apple"))   # ['apple']
print(add_item("banana"))  # ['apple', 'banana'] — unexpected!

# Fix: use None as default
def add_item(item, basket=None):
    if basket is None:
        basket = []
    basket.append(item)
    return basket