Future Skills · Learning hub
Machine Learning Learning Hub
Train models that learn from data — and know whether to trust them. 6 lessons, a Skill Check and guided projects — free.
- Free tutorials
- Practice tests
- Projects
- Lessons
- 6
- Reading time
- 73 min
- Skill Check
- 12 questions
- Projects
- 2
About this skill
Machine learning is how a bank flags fraud, how a food app predicts your delivery time and how a hospital spots risky scans early. This track takes you from "I have heard of ML" to training, evaluating and saving your own models with Python and scikit-learn. Every idea comes with intuition first, then a small piece of code you can run: preparing features, regression, classification, the metrics that tell you if a model is any good, and a complete end-to-end project.
Who it is for: Students who know basic Python and want to move into ML, data science or AI engineering, and freshers preparing for ML interview rounds.
After this track you will be able to
- Explain the machine learning workflow from raw data to prediction
- Prepare features: encoding, scaling and avoiding data leakage
- Train and interpret linear regression models
- Build classifiers with logistic regression, decision trees and random forests
- Choose the right metric: precision, recall, F1, RMSE and cross-validation
- Build, evaluate and save a complete scikit-learn pipeline
Learning path
Lessons, in order
Each lesson takes about 10 minutes and ends with a small task. 0 of 6 completed.
-
1ML FoundationsHow models learn from data, and how to prepare features that help them.
- How Machine Learning Works: Data, Features, Training and PredictionMachine learning is a recipe: examples go in, a model comes out, and the model makes predictions on new data. Learn every step of that recipe — features, labels, train/test split, training, overfitting — with one running example.
- Feature Engineering and Data Preparation for Machine LearningBetter features beat fancier algorithms. Learn how to encode categories, scale numbers, handle missing values, create useful new features — and avoid data leakage, the mistake that makes models look great until they meet real data.
-
2Core ModelsRegression and classification models you will use again and again.
- Linear Regression Explained: Predicting Numbers with a Straight LineLinear regression is the "hello world" of machine learning — and still used everywhere. Learn the intuition, what the coefficients mean, how the model learns by minimising error, and how to train and read one in Python.
- Classification Models: Logistic Regression, Decision Trees and Random ForestsSpam or not, fraud or genuine, pass or fail — classification is the most common ML task in business. Learn how the three workhorse classifiers think, their strengths and weaknesses, and how to train and compare them.
-
3Evaluating & Shipping ModelsMeasure a model honestly, then build a complete pipeline.
- Model Evaluation Metrics: Confusion Matrix, Precision, Recall, F1 and Cross-ValidationAccuracy alone can make a useless model look excellent. Learn the confusion matrix, precision, recall, F1, ROC-AUC, regression metrics and cross-validation — and how to pick the metric that matches the real cost of mistakes.
- Your First End-to-End Machine Learning Project with scikit-learnPut everything together: frame the problem, explore the data, build a leak-proof pipeline, compare models with cross-validation, tune, evaluate once, save the model and use it for predictions — the workflow interviewers want to hear.
Build
Machine Learning projects
Finish the lessons, then prove the skill with a project you can show.
FAQ
Machine Learning: common questions
How much maths do I need for machine learning?
To start: school-level algebra, averages and the idea of a slope. This track explains every concept with intuition and code before any formula. If you go deeper into research later, linear algebra, calculus and probability become important — but you can build useful models long before that.
Should I learn Python first?
Yes. You should be comfortable with variables, loops, functions, lists and dictionaries. The free Python track on this site covers exactly that; pandas basics from the Data Analytics track help too.
Machine learning or deep learning — what should I learn first?
Machine learning. Train/test splits, overfitting, features and evaluation apply to deep learning as well, and they are much easier to understand on small tabular datasets. Most business problems on tables are still solved best with the models in this track.
Is ML still worth learning now that LLMs exist?
Very much. LLMs are great with language, but predicting churn, prices, demand, fraud or risk from tables of data is still classic ML — and companies need people who can evaluate any model honestly, including LLM-based ones.
Preparing with friends?
Share this page with your study group or placement batch.