Installation and Environment Setup

Before we start building models, let's get your environment set up correctly. Scikit-learn is designed to work smoothly within the existing Python scientific computing stack. Having a proper setup ensures that you can follow along with the examples and apply these techniques to your own projects without compatibility issues.

As mentioned in the course prerequisites, you should already have Python installed, along with a basic familiarity with NumPy and Pandas. Scikit-learn builds directly on these libraries, particularly NumPy for its array structures and mathematical functions, and integrates well with Pandas DataFrames.

Using Virtual Environments (Recommended)

It's highly recommended to use a virtual environment for your Python projects, including this course. Virtual environments isolate project dependencies, preventing conflicts between different projects that might require different versions of the same library.

Using venv (Python's built-in tool):

1. Create an environment: Open your terminal or command prompt, navigate to your project directory, and run:

   python -m venv sklearn-env
   

   (Replace sklearn-env with your preferred environment name).

2. Activate the environment:

   • On macOS and Linux:

     source sklearn-env/bin/activate
     

   • On Windows:

     sklearn-env\Scripts\activate
     

   Your terminal prompt should now indicate that the virtual environment is active.

Using conda (if you use Anaconda or Miniconda):

1. Create an environment:

   conda create --name sklearn-env python=3.9
   

   (Replace sklearn-env with your chosen name and specify a Python version if desired). Conda will ask you to confirm the packages to be installed.

2. Activate the environment:

   conda activate sklearn-env
   

Working within an active virtual environment ensures that packages you install are contained and won't interfere with your global Python installation or other projects.

Installing Scikit-learn

With your virtual environment activated, you can install Scikit-learn. The library itself depends on NumPy and SciPy. Installing Scikit-learn using standard package managers like pip or conda typically handles these dependencies automatically.

Using pip:

This is the standard Python package installer. It's usually the simplest method if you're not using the Anaconda distribution.

pip install scikit-learn

This command will download and install the latest stable version of Scikit-learn, along with required versions of NumPy and SciPy if they are not already present in your environment.

Using conda:

If you are using Anaconda or Miniconda, conda is the preferred package manager.

conda install scikit-learn

Conda manages packages and environments, including potentially complex binary dependencies, which can sometimes be beneficial, especially on Windows.

Core Dependencies

Scikit-learn's main dependencies are:

• NumPy: For efficient array operations, linear algebra, and numerical computing (<https://numpy.org/>). Scikit-learn's primary data structures are often NumPy arrays.
• SciPy: Provides fundamental algorithms for scientific computing, which Scikit-learn uses for some of its implementations (<https://scipy.org/>).

These libraries are foundational to the scientific Python ecosystem and are almost always installed alongside Scikit-learn.

Optional Dependencies for Enhanced Functionality

While not strictly required for basic Scikit-learn usage, you'll often work with other libraries in a typical machine learning workflow:

• Pandas: Essential for data manipulation and analysis, particularly for loading and preparing data stored in tabular formats like CSV files (<https://pandas.pydata.org/>). Scikit-learn integrates very well with Pandas DataFrames.
• Matplotlib: The standard library for creating static, animated, and interactive visualizations in Python (<https://matplotlib.org/>). Useful for exploring data and evaluating models.
• Seaborn: Built on top of Matplotlib, Seaborn provides a high-level interface for drawing attractive and informative statistical graphics (<https://seaborn.pydata.org/>).

If these are not already installed in your environment, you can add them using pip or conda:

Using pip:

pip install pandas matplotlib seaborn jupyterlab

(We've included jupyterlab here as it provides a convenient interactive environment for data science work, which you might find useful).

Using conda:

conda install pandas matplotlib seaborn jupyterlab

Verifying the Installation

To confirm that Scikit-learn is installed correctly, you can start a Python interpreter or a Jupyter Notebook session within your activated virtual environment and run the following commands:

import sklearn
import numpy
import scipy
import pandas # Optional, if installed
import matplotlib # Optional, if installed

# Print the scikit-learn version
print(f"Scikit-learn version: {sklearn.__version__}")
print(f"NumPy version: {numpy.__version__}")
print(f"SciPy version: {scipy.__version__}")
# print(f"Pandas version: {pandas.__version__}") # Uncomment if installed
# print(f"Matplotlib version: {matplotlib.__version__}") # Uncomment if installed

If these commands execute without raising an ImportError and print the library versions, your environment is set up correctly, and you're ready to proceed with exploring Scikit-learn's features. We will perform a more formal verification in the hands-on practical at the end of this chapter.