Machine Learning Basics
Understand what ML is, how it works, and train real models in your browser, no install required.
Intermediate | 8 Modules | scikit-learn in Browser | ND FriendlyBefore You Start: Prerequisites
- Completed Introduction to Python or Python Practice Labs (or equivalent experience)
- Comfortable with variables, lists, loops, and functions in Python
- No math beyond high school algebra required, we explain the rest as we go
- A modern browser (Chrome or Firefox recommended), no installs, no accounts
What You’ll Be Able to Do
- Explain what machine learning is (and isn’t) in plain English
- Choose the right type of ML for a given problem
- Clean and prepare data for a model
- Train and evaluate classification and regression models using scikit-learn
- Interpret results, read a confusion matrix, and understand precision vs recall
- Build a complete ML pipeline from scratch: load → clean → train → evaluate → interpret
Course Modules
What IS Machine Learning?
Rules based vs learning based programming. The 3 types of ML. The ML workflow from data to prediction.
Your First Dataset
What is a dataset? Rows, columns, features, labels. Exploring data and understanding what you’re working with.
Preparing Data
Cleaning messy data, encoding categories, scaling features, and splitting into train/test sets.
Your First Classifier: KNN
K Nearest Neighbors: how distance based classification works. Fit, predict, score with scikit-learn.
Decision Trees
How trees ask yes/no questions to classify data. Tree depth, Gini impurity, overfitting, and visualization.
Linear Regression
Predicting numbers instead of categories. Line of best fit, MSE, R² score, and when regression applies.
Evaluating Models
Why accuracy alone can mislead. Confusion matrices, precision, recall, F1 score, and when each matters.
Capstone Project
Build a complete ML pipeline. Load, explore, clean, train three models, compare results, and present your findings.
A Note on Pyodide Load Times
Each module uses Pyodide to run Python and scikit-learn directly in your browser. The first load takes about 15 to 20 seconds depending on your connection. After that, code runs instantly. Be patient on the first run, it’s worth it!