Setting Up Your ML Environment

30 min readvideoPython Foundations for ML

1 of 24Python for Machine Learning

Setting Up Your ML Environment

Before writing a single line of machine-learning code you need a reliable, reproducible Python environment. This lesson walks you through every step: choosing a Python distribution, managing packages, and launching your first Jupyter notebook.

1. Installing Python 3.11+

Machine-learning libraries are moving fast. Python 3.11 brought a significant speed-up (10-60 % on many benchmarks), and 3.12 improved error messages. Always install the latest stable release from python.org or use a distribution such as Anaconda / Miniconda.

Verify your installation in a terminal:

python

# Check Python version
python --version          # Should show 3.11.x or higher

# Check pip is available
pip --version

2. Conda vs Pip — Which Package Manager?

Both tools install Python packages, but they solve different problems. The table below highlights the key differences.

Feature	pip	conda
Package source	PyPI (Python Package Index)	Anaconda / conda-forge channels
Language support	Python only	Python, R, C/C++ libraries, CUDA toolkits
Dependency solver	Basic (pip 23+ improved)	SAT solver — resolves entire environment
Virtual environments	`python -m venv` (lightweight)	`conda create -n env` (heavier, more isolated)
Binary packages	Wheels (platform-specific)	Pre-compiled for each OS/arch
Speed	Fast installs	Slower solving, but `mamba` helps
Best for	Lightweight projects, CI/CD	Data-science stacks, GPU drivers

3. Creating a Virtual Environment

A virtual environment isolates your project's dependencies from the system Python and from other projects. Never install ML packages into your base/system Python.

Option A — conda

python

# Create an environment named "ml-course" with Python 3.11
conda create -n ml-course python=3.11 -y

# Activate it
conda activate ml-course

Option B — venv (built-in)

python

# Create the environment
python -m venv ml-course-env

# Activate (Linux / macOS)
source ml-course-env/bin/activate

# Activate (Windows PowerShell)
ml-course-env\Scripts\Activate.ps1

4. Installing the Core ML Stack

With your environment activated, install the packages we will use throughout this course:

python

# Core scientific computing
pip install numpy pandas matplotlib

# Machine learning
pip install scikit-learn

# Jupyter notebooks
pip install jupyterlab notebook

# Optional but handy
pip install seaborn tqdm ipywidgets

5. Verifying Your Installation

Run this quick check script to make sure everything is installed correctly.

verify_install.py Run

python

import sys
print(f"Python  : {sys.version}")

import numpy as np
print(f"NumPy   : {np.__version__}")

import pandas as pd
print(f"Pandas  : {pd.__version__}")

import matplotlib
print(f"Matplotlib: {matplotlib.__version__}")

import sklearn
print(f"Scikit-learn: {sklearn.__version__}")

print("\n All core packages installed successfully!")

Python  : 3.11.7 (main, Dec  8 2023, 14:22:46)
NumPy   : 1.26.2
Pandas  : 2.1.4
Matplotlib: 3.8.2
Scikit-learn: 1.3.2

 All core packages installed successfully!

6. Jupyter Notebook Basics

Jupyter notebooks (.ipynb) let you mix code, output, visualisations, and Markdown text in a single document. They are the de-facto standard for exploratory ML work.

python

# Launch JupyterLab (recommended)
jupyter lab

# Or classic notebook
jupyter notebook

Key keyboard shortcuts inside a notebook:

Shortcut	Action
`Shift + Enter`	Run cell, move to next
`Ctrl + Enter`	Run cell, stay in place
`Esc` then `A`	Insert cell above
`Esc` then `B`	Insert cell below
`Esc` then `M`	Convert cell to Markdown
`Esc` then `DD`	Delete cell

7. IDE Recommendations

For writing .py scripts and modules you'll want a good editor. Here are the top choices for ML work:

IDE / Editor	Strengths	Free?
VS Code	Excellent Python extension, integrated terminal, Jupyter support, Git	Yes
PyCharm Professional	Advanced refactoring, scientific mode, database tools	Free for students
JupyterLab	Best for notebooks, built-in file browser, terminal	Yes
Google Colab	Free GPU, zero setup, shareable	Yes (free tier)

"The best environment is the one you actually use consistently. Pick one, learn its shortcuts, and stick with it."

Up next · Python Refresher: Data Types and Control Flow