Machine Learning for Everyone
Build a sound prediction workflow from problem definition and data splitting to model comparison and monitoring.
Before you start: Basic arithmetic and familiarity with tables; Python is optional for the core explanations.
Start Lesson 1
Your module quizzes
Five multiple-choice questions after each module. Get a percentage and feedback, review explanations, and retry anytime. Aim for 80%; quizzes never lock lessons.
Sign in to save your personal results, or practice as a guest.
20 modules, 80 lessons
Follow the modules in order, or open any lesson to review a specific concept. Every lesson includes reading, practice, a quick check, and a transcript. Published videos include optional English captions.
Problem and Data
Lessons 1–4 · 4/4 videos available
Turn a Real Problem into a Learning Task
A delivery dispatcher wants fewer late arrivals.
Training, Validation, and Test Data
Our dispatcher has a thousand past deliveries and a clear task: flag a likely delay before departure.
Explore Data Before Fitting a Model
A model can fit a spreadsheet full of mistakes without telling you which mistakes matter.
Preprocessing Without Data Leakage
Standardizing a feature sounds harmless, but even preprocessing can leak information.
Module 1 quiz
5 questions · score and feedback · retry anytime
Core Prediction Models
Lessons 5–8 · 4/4 videos available
Linear Regression as a Useful Baseline
A straight-line model can be a strong starting point even when the real world is complicated.
Logistic Regression and Decision Thresholds
A package is either late or on time, but the evidence before departure is uncertain.
Nearest Neighbors: Learn from Similar Examples
You want to estimate a home's price by looking at similar homes nearby.
Decision Trees: Follow a Sequence of Questions
A decision tree predicts by asking a sequence of questions.
Module 2 quiz
5 questions · score and feedback · retry anytime
Model Selection and Evaluation
Lessons 9–12 · 4/4 videos available
Random Forests: Combine Different Trees
One decision tree can change dramatically when a few training examples change.
Boosting: Improve a Model in Stages
A first model makes systematic mistakes.
Regularization and Cross-Validation
A flexible model can explain the training data in many ways, including ways that follow noise.
Evaluate Imbalanced Classification Honestly
Only one out of every hundred manufactured parts is defective.
Module 3 quiz
5 questions · score and feedback · retry anytime
Unsupervised Learning and Deployment
Lessons 13–16 · 4/4 videos available
Clustering: Find Groups Without Target Labels
A shop has many customer records but no predefined customer categories.
PCA: Compress Features into Directions
Several measurements may tell us nearly the same thing.
Deploy and Monitor a Complete Prediction Pipeline
A model that worked yesterday can fail today even when its code has not changed.
Capstone: Build and Compare a Prediction System
Now let us assemble a complete learning project: predict delivery delay before departure.
Module 4 quiz
5 questions · score and feedback · retry anytime
Learning Concepts from Examples
Lessons 17–20 · 4/4 videos available
Describe a Hypothesis Space
A dispatcher proposes a rule using route, weather, and departure hour.
Specific and General Rules
A rule flags one particular rainy route but misses other late deliveries.
Version Spaces and Conflicting Examples
Two records with the same known inputs have conflicting delay labels.
Inductive Bias: Why Examples Are Not Enough
The team must decide which routes to compare before looking at all future trips.
Module 5 quiz
5 questions · score and feedback · retry anytime
Data Quality and Sampling
Lessons 21–24 · 4/4 videos available
Find Missing and Incorrect Labels
Some deliveries are labeled late even though arrival records are missing.
Recognize Sampling Bias
Training mostly contains downtown routes while deployment includes rural trips.
Handle Rare Outcomes with Care
Only a few deliveries are more than an hour late.
Measure What Is Known at Prediction Time
A late-arrival field appears highly predictive in historical data.
Module 6 quiz
5 questions · score and feedback · retry anytime
Decision Trees in Depth
Lessons 25–28 · 4/4 videos available
Measure Impurity in a Split
A tree splits past deliveries into route groups.
Choose a Feature with Information Gain
Distance or weather could be the tree's first question.
Prune a Tree Before It Memorizes Noise
A deep tree creates one leaf for nearly every unusual trip.
Explain a Path Without Overclaiming
A dispatcher asks why one delivery was flagged.
Module 7 quiz
5 questions · score and feedback · retry anytime
Probability and Bayes
Lessons 29–32 · 4/4 videos available
Prior Beliefs and New Evidence
The service has a historical late-delivery rate before today's forecast arrives.
Bayes' Rule with a Worked Example
A weather alert catches many late deliveries but also some on-time ones.
Naive Bayes and Its Assumptions
The team scores weather, route, and distance together.
Check Whether Probabilities Are Calibrated
Among deliveries given a 70% late risk, far fewer are actually late.
Module 8 quiz
5 questions · score and feedback · retry anytime
Reliable Model Evaluation
Lessons 33–36 · 4/4 videos available
Confusion Matrix for Real Decisions
The owner wants to see missed delays and unnecessary warnings separately.
Cross-Validation with Groups and Time
Several records share a driver and repeated route conditions.
Uncertainty in an Estimated Score
A test accuracy changes when only a few deliveries are added.
Compare Two Models Without Test-Set Tuning
Two models differ by one percentage point on the held-out period.
Module 9 quiz
5 questions · score and feedback · retry anytime
Linear Prediction
Lessons 37–40 · 0/4 videos available
Fit a Line by Minimizing Error
The dispatcher needs a numeric estimate of minutes late.
Read Coefficients with Context
A positive distance coefficient appears in the fitted model.
Regularize a Flexible Linear Model
Many related route indicators make coefficients unstable.
Choose a Classification Threshold
The dispatcher can review only a limited number of warnings.
Module 10 quiz
5 questions · score and feedback · retry anytime
Instance-Based Learning
Lessons 41–44 · 0/4 videos available
Define a Distance That Makes Sense
Two deliveries have equal mileage but different road conditions.
Scale Features Before Nearest Neighbors
Mileage dominates an unscaled neighbor distance over a small weather score.
Tune Neighbor Count on Validation Data
One-neighbor predictions fluctuate wildly on isolated deliveries.
Recognize High-Dimensional Distance Problems
The dataset adds dozens of weakly relevant delivery features.
Module 11 quiz
5 questions · score and feedback · retry anytime
Ensembles
Lessons 45–48 · 0/4 videos available
Bootstrap Samples and Bagging
The team trains several trees on different samples of deliveries.
Random Forest Feature Choices
Every tree keeps choosing mileage first.
Boosting as Sequential Correction
One model struggles with delayed trips from bad weather.
Inspect Shared Errors in an Ensemble
All ensemble members miss the same rural route.
Module 12 quiz
5 questions · score and feedback · retry anytime
Margins and Kernels
Lessons 49–52 · 0/4 videos available
Separate Classes with a Margin
Late and on-time trips overlap in feature space.
Soft Margins and Difficult Points
A few mislabeled trips lie on the wrong side of the boundary.
A Kernel as a Similarity Rule
A straight boundary fails to separate two route patterns.
Compare Linear and Nonlinear Boundaries
The curved boundary looks much better on training deliveries.
Module 13 quiz
5 questions · score and feedback · retry anytime
Neural Models in Context
Lessons 53–56 · 0/4 videos available
Perceptrons as Simple Learners
A one-layer learner gives a simple yes-or-no delivery warning.
Gradient Updates and Loss
The model's warning score is too high for an on-time delivery.
When a Neural Model Helps
A tabular delay baseline already performs well.
Compare with the Neural Networks Course
The team wants to connect this tabular task with neural-network ideas.
Module 14 quiz
5 questions · score and feedback · retry anytime
Unsupervised Patterns
Lessons 57–60 · 0/4 videos available
Choose a Clustering Question
A dispatcher wonders whether routes form natural groups without delay labels.
Run and Inspect K-Means
The team asks k-means to form three route groups.
Hierarchical Groups and Their Limits
Some routes appear nested within broader regional groups.
PCA for Compact Representations
Dozens of correlated route indicators make the dashboard crowded.
Module 15 quiz
5 questions · score and feedback · retry anytime
Rules and Explanations
Lessons 61–64 · 0/4 videos available
Learn If-Then Rules
Operations staff write a rule for rainy long routes.
Precision and Coverage of a Rule
A warning rule catches many late trips but flags too many on-time ones.
Combine Rules Without Hidden Conflicts
Two rules both apply to the same trip but disagree.
Use Prior Knowledge Carefully
A dispatcher knows that a certain bridge closes every spring.
Module 16 quiz
5 questions · score and feedback · retry anytime
Learning Theory
Lessons 65–68 · 0/4 videos available
Training Fit Versus Generalization
A model nearly memorizes every training delivery.
Sample Size and Hypothesis Complexity
The team has a small dataset and many flexible candidate rules.
PAC Learning as a Thought Model
A colleague asks whether learning can be guaranteed from examples.
Recognize When the World Changes
A highway closes and travel patterns change suddenly.
Module 17 quiz
5 questions · score and feedback · retry anytime
Reinforcement Learning
Lessons 69–72 · 0/4 videos available
Actions, States, and Rewards
The dispatcher may warn, reroute, or wait.
Delayed Feedback in a Markov Decision Process
A reroute saves time today but affects later driver availability.
Estimate Values from Experience
The dispatcher observes a few outcomes after each route choice.
Balance Exploration with Safe Decisions
A new route might be faster but has little evidence.
Module 18 quiz
5 questions · score and feedback · retry anytime
Responsible Operations
Lessons 73–76 · 0/4 videos available
Audit Leakage Before Deployment
An unexpectedly predictive field appears in the deployment pipeline.
Evaluate Errors Across Groups
One district receives many unnecessary delay alerts.
Monitor Shift and Calibration
The live score distribution shifts after a schedule change.
Retrain and Roll Back with Evidence
A replacement model beats the old one in one week.
Module 19 quiz
5 questions · score and feedback · retry anytime
Final Machine-Learning Project
Lessons 77–80 · 0/4 videos available
Frame a Prediction and Baseline
The final project starts with an operational delay question.
Split and Train Reproducibly
The team needs another person to repeat its training run.
Compare Candidates on Honest Data
Several candidate models look close on validation data.
Report Limits and a Deployment Plan
Operations asks whether the final model is safe to try.
Module 20 quiz
5 questions · score and feedback · retry anytime
References and further reading
These books guide the topic selection and provide deeper study. ToolsInVision's explanations, worked examples, and exercises are original teaching material; the books are not reproduced here.
Machine Learning — Tom M. Mitchell. Further study of learning tasks, decision trees, evaluation, Bayesian learning, and generalization.