Anaconda Distribution Setup

What is Anaconda?

Anaconda is a Python (and R) distribution designed for data analytics, data science, and machine learning. Instead of installing Python and dozens of libraries manually, Anaconda bundles:

• A Python interpreter
• The conda package/environment manager
• Popular data-science libraries (often preinstalled)
• Optional GUI tools like Anaconda Navigator

Using Anaconda gives you two big benefits:

1. Environment isolation (separate projects with separate dependencies)
2. Reproducible installs (others can recreate your environment)

Note

Anaconda and Miniconda

• Anaconda: larger download, includes many packages and Navigator.
• Miniconda: smaller, installs only Python + conda; you install packages you need.

If you have limited disk space, Miniconda is often best.

Choosing Anaconda vs standard Python + pip

You can do data analytics using standard Python and pip, but for beginners and many data workflows, Anaconda is great because:

• Installing compiled libraries (NumPy/Pandas) is easier
• You avoid many “wheel/build tools” issues
• You can combine packages from conda + pip when needed

Step 1: Download Anaconda

1. Open the official site: https://www.anaconda.com/download
2. Choose your OS (Windows/macOS/Linux)
3. Download the Python 3.x installer

Caution

Only download from the official website. Avoid random “download” mirrors.

Step 2: Install Anaconda

Windows installation tips

• During installation, you may see these options:
  • Add Anaconda to PATH: usually not recommended (can conflict with other Python installs)
  • Register Anaconda as default Python: recommended if you want Anaconda to be your main Python

After installation, you’ll typically use:

• Anaconda Prompt (recommended)
• Anaconda Navigator (GUI)

macOS installation tips

Use the .pkg installer and follow the prompts.

After installation:

• Use Terminal with conda after initializing your shell

Linux installation tips

Typically you install via a .sh script.

After installation you may need:

• Initialize conda for your shell
• Restart terminal

Step 3: Verify installation

Open Anaconda Prompt (Windows) or Terminal (macOS/Linux) and run:

command

conda --version

You should see a version like:

text

conda 24.x.x

Also confirm Python:

command

python --version

Step 4: conda basics you must know

What is a conda environment?

A conda environment is a folder containing:

• A specific Python version
• Installed libraries for one project

This prevents one project’s dependencies from breaking another project.

Create a new environment

command

conda create -n data-analytics python=3.12

Activate the environment

• Windows:

command

conda activate data-analytics

• macOS/Linux:

command

conda activate data-analytics

(Activation command is the same, but your prompt will look different.)

List environments

command

conda env list

Install packages

command

conda install numpy pandas matplotlib seaborn jupyter

Update packages

command

conda update numpy

Remove an environment

command

conda env remove -n data-analytics

Step 5: Using Anaconda Navigator

Anaconda Navigator is a GUI app that makes it easy to:

• Launch Jupyter Notebook / JupyterLab
• Install packages
• Manage environments

Common workflow:

1. Open Navigator
2. Choose an environment (top dropdown)
3. Launch Jupyter

Visualize it

This flowchart traces the full setup path, from downloading the installer to having a ready-to-use analytics environment.

diagram Anaconda install flow mermaid

  flowchart TD
  A["Download installer"] --> B["Run installer (Windows/macOS/Linux)"]
  B --> C["Open Anaconda Prompt / Terminal"]
  C --> D["Verify: conda --version"]
  D --> E["conda create -n data-analytics python=3.12"]
  E --> F["conda activate data-analytics"]
  F --> G["conda install numpy pandas matplotlib jupyter"]
  G --> H["Ready to launch Jupyter"]

From downloading the installer to a working, activated data-analytics environment

From the book

McKinney’s own setup recipe (using Miniconda) is worth copying: configure conda-forge as your default channel and set channel_priority strict before installing anything (conda config --add channels conda-forge, conda config --set channel_priority strict). This one-time setup step avoids many confusing dependency conflicts later. He also warns that you must run conda activate every time you open a new terminal — forgetting this is one of the most common “why isn’t my package found?” mistakes.

Common issues & fixes

Problem: conda: command not found

This usually means conda isn’t initialized for your shell.

Try:

command

conda init

Then close and reopen your terminal.

Problem: installing is slow

Try updating conda and using a faster solver:

command

conda update conda

You can also consider mamba later (a faster conda alternative).

Problem: conflicting packages

Create a new clean environment for each project. That’s the point of conda.

Next

Continue to: Jupyter Notebook Interface to learn how to create notebooks and run data analytics interactively.

🧪 Try It Yourself

Exercise 1 – Build a conda create Command

pch.exerciseTag pch.exerciseTitle

Exercise 2 – Parse conda --version Output

pch.exerciseTag pch.exerciseTitle

Exercise 3 – List Environments From a Dictionary

pch.exerciseTag pch.exerciseTitle