Menu
CoddyTech
START LEARNING
flag En iconEnglishdown icon
machine Learning iconPart of the Machine Learning JourneyNewJourney

Unsupervised Learning in Python

Find the groups, the shape and the odd ones out in data that has no labels. Cluster, shrink and scan real datasets with scikit-learn, and see every result on a chart.

START LEARNING(it's free)
View syllabusUnsupervised Learning playground

1,000+ codders enrolled

51 lessons51 challenges210 quiz questions

  • Beginner friendly
  • sparkles iconAI-assisted coding help
  • hint iconHands-on interactive lessons
  • quiz iconQuizzes to test your knowledge
  • certificate iconFree certificate of completion

Syllabus

1 sections51 lessons51 challenges210 quiz questions

This section is part of the Machine Learning Journey. The full syllabus has more sections - click any preview below to view it on the Journey page.

  1. Section 1Data and Plots60 lessons
  2. Section 2Statistics for ML64 lessons
  3. Section 3Supervised Learning58 lessons
  4. Section 4

    Unsupervised Learning

    Start sectionStartExpandCollapse    Find structure in data without labels: k-means, hierarchical and density clustering, PCA to see and compress many features, and anomaly detection, ending with a digits exploration project.51 lessons51 challenges210 questions

    Finding Groups

    5 lessons520

    Choosing k

    5 lessons520

    Reading Clusters

    5 lessons520

    Hierarchical Clustering

    4 lessons415

    Density Clustering

    4 lessons415

    Fewer Dimensions

    5 lessons520

    PCA in Practice

    5 lessons520

    Anomaly Detection

    5 lessons520

    Clusters at Work

    4 lessons415

    Project: Digit Explorer

    5 lessons525

    Final Challenges

    4 lessons420
View the full Machine Learning Journey syllabus
What you get
Everything you'll use to learn to code

Learn by Doing

Write real code, query databases, build websites, and master AI prompts. Our interactive lessons cover every skill modern developers need.

playground.js
Code Editor
1const greeting = "Hello, Coddy!"
2function sayHi(name) {
3    return greeting + " " + name
4}
5
bottombar Collapse icon
Test #1test Case Success icon
Test #2test Case Success icon
Test #3test Case Failure icon
Input
"Alex"
Output
"Hello, Coddy! Alex"

Build Your Streak

Stay consistent and watch your progress grow! Track your daily coding habit, protect your streak with freeze days, and earn rewards for showing up every day.

12 days streak

Return tomorrow to keep your streak!

fire Filled icon
left icon

January 2026

right icon

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

21

22

23

24

25

26

27

28

29

30

product Double Or Nothing icon

Double or Nothing

Day 5 of 7

fire Freeze icon

Streak Freeze

2 left

Code Anywhere

Take your coding journey on the go! No setup, no downloads - just open and start coding. Available on iOS, Android and Web with 4.9 star ratings.

Python
7Streak
250Score
5Energy
Variables
journey Hex Done Base iconjourney Hex Done Shadow iconjourney Hex Done Top iconjourney Lesson Done icon
journey Path Right Done icon
journey Hex Done Base iconjourney Hex Done Shadow iconjourney Hex Done Top iconjourney Lesson Done icon
journey Path Left Done icon
journey Hex Active Base iconjourney Hex Active Shadow iconjourney Hex Active Top iconjourney Lesson Theory Challenge icon
CONTINUE
journey Path Right icon
journey Hex Locked Base iconjourney Hex Locked Shadow iconjourney Hex Locked Top iconjourney Lesson Theory Challenge icon
journey Path Left icon
journey Hex Locked Base iconjourney Hex Locked Shadow iconjourney Hex Locked Top iconjourney Lesson All icon
Journey
Goals
Leaderboard
Profile
4.9
StarStarStarStarStar
Rating

You're Not Alone

Compete on global leaderboards, invite friends to earn rewards, and celebrate each other's wins. Coding is better with friends!

Challenger League
Challenger LeagueTop 7 advance
leaderboard First icon1
avatar 1 icon
fire Filled icon
Alex7+ Days
2840
leaderboard Second icon2
avatar 2 icon
fire Filled icon
Jordan7+ Days
2650
leaderboard Third icon3
avatar 3 icon
fire Filled icon
Sam7+ Days
2420
4
avatar 4 icon
Casey
2180
5
avatar placeholder icon
fire Filled icon
Morgan7+ Days
1950
leaderboard Arrow Up iconPromotion zoneleaderboard Arrow Up icon

Every way to learn

Read, listen, test yourself, ask the AI, or look up anything you've already covered. Every lesson meets you where you are.

Intro to Variables
Audio

A variable is a named container that stores a value you can reference later in your program.

In Python, you create one by writing the name, an equals sign, then the value you want to store.

The value can change over time - reassigning the name simply points it to a new value.

1xSarah

Earn a Certificate

Earn certificates for every course you complete. Add them to your LinkedIn profile and resume to showcase your coding expertise to employers.

CoddyCertificate of Completion
This certifies thatJohn Doehas successfully completed
python iconPython Fundamentals
Verified
DateJan 2026
LinkedInAdd to LinkedIn

Why learn unsupervised learning with Coddy

  • Clustering you can see. Watch k-means move its centers, read a dendrogram's merge heights, and see DBSCAN find moon-shaped groups that k-means cuts in half. Every lesson has a chart, and yours appears in the Plots tab as you run.
  • The core methods: k-means, choosing k with the elbow and silhouette scores, hierarchical clustering, DBSCAN, PCA and explained variance, isolation forests and local outlier factor, all with scikit-learn.
  • Results you can explain. You profile each cluster in real units, compare clusters with known labels, and name the segments a business would act on, instead of stopping at a list of cluster numbers.
  • 64 dimensions made visible: handwritten digits projected onto a flat map, PCA components drawn as images, and pictures rebuilt from a few numbers.

Frequently asked questions about unsupervised learning

What is unsupervised learning?

Unsupervised learning finds structure in data that has no labels. Nobody tells the model the right answer; it looks for groups of similar rows (clustering), for a few directions that keep most of the information (dimensionality reduction), or for rows unlike the rest (anomaly detection). It is how you explore a new dataset, segment customers or spot fraud before anyone has labelled it.

What are examples of unsupervised learning?

Grouping shoppers by what they buy, reducing an image's colours to the eight that matter, mapping 64-pixel digit images onto a flat chart, finding unusual server readings, and splitting cell measurements into two kinds without being told which is which. The course builds each of these.

Which algorithms does the course cover?

k-means with the elbow method and silhouette scores, agglomerative (hierarchical) clustering with dendrograms, DBSCAN, PCA, isolation forests and local outlier factor. You also learn why scaling comes first and when each method fails, so you can pick the right one.

How do I choose the number of clusters?

There is rarely one right answer, so the course teaches three ways to look: the elbow of the inertia curve, silhouette scores, and a dendrogram's merge heights. Then you check the clusters in real units and ask whether a person could name them. A k that gives groups nobody can describe is the wrong k.

Should I learn supervised learning first?

It helps. The supervised learning section teaches scaling, pipelines and the scikit-learn workflow that this section reuses, and its distance ideas from k-nearest neighbors return here in clustering. If you already train models with scikit-learn, you can start here.

How long does the unsupervised learning section take?

51 lessons, about 8 to 9 hours at your own pace: a few weeks at half an hour a day, including the digit explorer project and four final challenges.

Can I learn unsupervised learning for free?

Yes. The lessons, the charts, the AI hints and the section certificate are free. A free account works through a few lessons a day, and Pro removes the daily limit.

Learning paths that include this course

This course is one step of a longer path: several courses in the order a mentor would teach them, each with its own free certificate.

Coddy programming languages illustration

Learn Unsupervised Learning with Coddy

GET STARTED