Course 1 · Machine Learning

Machine Learning Fundamentals

Continue from the AI map in Course 0 and learn the loop that turns data into a model, evaluates it, and keeps it useful in production.

For
Learners who finished Course 0 or already understand the basic AI map
Level
Beginner 1–2/5
Full course
~120 min
Available now
1/12 lessons · 10 min

By the end

What you will be able to do

  • Explain the universal ML loop: data → learn → predict → evaluate.
  • Choose classification, regression, or clustering for a simple problem.
  • Separate train, validation, and test data and recognize overfitting.
  • Interpret basic accuracy, precision, and recall results.
  • Distinguish training from inference and describe the deploy, monitor, retrain lifecycle.

Before you start

Prerequisites

Saved on this device

Your progress

0 of 1 available lessons

Saved only in this browser. No account is required and no progress data leaves your device.

Start the first lesson →

Recommended path

Learn in order

Start with the lessons available today. New lessons will slot into this path without changing existing URLs.

  1. 01

    The ML Loop: Data → Learn → Predict → Evaluate

    See the core loop shared by every machine-learning model.

    10 min In development
    In development
  2. 02

    Feature, Label, and Target

    Separate what a model sees from what it needs to predict.

    10 min In development
    In development
  3. 03

    Classification: Predict a Category

    Use it when the answer is a class, such as spam or not spam.

    10 min In development
    In development
  4. 04

    Regression: Predict a Number

    Use it for continuous values such as price, time, or temperature.

    10 min In development
    In development
  5. 05

    Clustering: Let Data Form Groups

    Find structure in data that has no answer labels yet.

    10 min In development
    In development
  6. 06

    Why Split Train, Validation, and Test Data?

    Separate learning, tuning, and final evaluation for an honest result.

    10 min In development
    In development
  7. 07

    Overfitting: Memorizing Instead of Learning

    Recognize when a model memorizes training data too closely.

    10 min In development
    In development
  8. 08

    Accuracy, Precision, and Recall

    Match the metric to the mistakes that matter in the real task.

    10 min In development
    In development
  9. 09

    What Is a Neural Network?

    Meet layers, weights, and activations before moving into deep learning.

    10 min In development
    In development
  10. 10

    Training vs Inference

    Separate the phase where a model learns from the phase where it serves predictions.

    10 min Available
    Open lesson
  11. 11

    Gradient and Learning Rate Without the Math Fear

    See how small weight updates move a model toward lower error.

    10 min In development
    In development
  12. 12

    From Notebook Model to a Real Product

    Connect deployment, monitoring, and retraining into one lifecycle.

    10 min In development
    In development

Course 1 outcome

Follow a model from raw data through learning, evaluation, deployment, and monitoring.

When all twelve lessons are available, you will be able to frame a problem, split data, read core metrics, and explain why deployed models need monitoring and retraining.