- 01Up nextAI, ML, Deep Learning and Neural Networks1.17 min20 XPCurrent
- 02How a Machine Actually Learns1.28 min20 XPOpen
- 03Supervised vs Unsupervised Learning1.39 min20 XPOpen
- 04Where ML Already Shows Up in Your Day1.48 min20 XPOpen
- 05The ML Lifecycle: From a Question to a Working Model1.510 min20 XPOpen
- 06Features, Targets, and What a Model Really Is1.610 min20 XPOpen
The learn track
Machine learning,
from zero.
Every idea explained in plain language, in the order it makes sense. Real examples, code you can run in Colab, and no jargon before it has been explained.
8 chapters/56 lessons/learn at your pace
Your progress
0/56lessons complete
0 of 1590 XP earned· Sign in to save progress
Next up: AI, ML, Deep Learning and Neural Networks
0% completeKeep going
The curriculum
A clear path, one lesson at a time.
Start anywhere, or follow the path in order. Every lesson is open to you.
- 01Why Python Became the Language of ML2.19 min20 XPOpen
- 02Jupyter and Google Colab: Where ML Code Is Written2.29 min20 XPOpen
- 03NumPy: Why Fast Arrays Matter2.312 min20 XPOpen
- 04pandas: Your Data as a Table2.413 min20 XPOpen
- 05Matplotlib: Seeing Your Data2.512 min20 XPOpen
- 06scikit-learn: A Toolbox of Ready-Made Models2.612 min20 XPOpen
- 07Just Enough Statistics2.714 min20 XPOpen
- 01What a Dataset Is, and Where to Find Good Ones3.19 min20 XPOpen
- 02Looking at Your Data Before Touching It3.210 min20 XPOpen
- 03Missing Values and Messy Data3.311 min20 XPOpen
- 04Turning Categories Into Numbers3.412 min20 XPOpen
- 05Feature Scaling3.511 min20 XPOpen
- 06Feature Engineering3.610 min20 XPOpen
- 07Train/Test Split3.710 min20 XPOpen
- 01Linear Regression: Finding the Best Line4.112 min20 XPOpen
- 02How Wrong Is the Line? Cost and MSE4.215 min25 XPOpen
- 03Gradient Descent: Walking Downhill4.317 min30 XPOpen
- 04Multiple Linear Regression: Learning from Several Clues4.418 min35 XPOpen
- 05Polynomial Regression: When the Relationship Bends4.517 min35 XPOpen
- 06Measuring a Regression Model: MAE, MSE, RMSE, R²4.619 min35 XPOpen
- 01Overfitting and Underfitting: When Generalisation Breaks5.117 min30 XPOpen
- 02Bias vs Variance: Two Ways a Model Can Miss5.216 min30 XPOpen
- 03Regularisation: Ridge and Lasso5.319 min35 XPOpen
- 04Validation Sets and Cross-Validation: Choosing Honestly5.418 min35 XPOpen
- 05Hyperparameters and Tuning: Choosing the Model Settings5.516 min30 XPOpen
- 06Data Leakage: When the Answer Sneaks Into the Features5.618 min40 XPOpen
- 01Classification vs Regression: Numbers or Categories?6.114 min25 XPOpen
- 02Logistic Regression: Turning Scores into Probabilities6.218 min35 XPOpen
- 03Confusion Matrix and Accuracy: Counting the Four Outcomes6.316 min30 XPOpen
- 04Precision, Recall, and F1: Which Mistakes Matter?6.417 min35 XPOpen
- 05Class Imbalance: When the Rare Case Matters Most6.516 min30 XPOpen
- 06ROC Curve and AUC: Ranking Across Thresholds6.617 min35 XPOpen
- 07K-Nearest Neighbours: Let Nearby Examples Vote6.716 min30 XPOpen
- 08Naive Bayes: Classify with Simple Probability Evidence6.817 min35 XPOpen
- 09Decision Trees: Ask a Sequence of Useful Questions6.917 min35 XPOpen
- 10Random Forest: Average Many Different Trees6.1017 min35 XPOpen
- 11Gradient Boosting: Add Small Trees One Correction at a Time6.1118 min40 XPOpen
- 12Support Vector Machines: Find a Boundary with Room Around It6.1218 min40 XPOpen
- 01Clustering vs Association: Groups or Things That Occur Together?7.112 min25 XPOpen
- 02K-Means: Assign Points, Move Centres, Repeat7.216 min35 XPOpen
- 03Hierarchical Clustering: Groups Inside Bigger Groups7.314 min30 XPOpen
- 04DBSCAN: Find Dense Groups and Leave Room for Noise7.416 min35 XPOpen
- 05PCA: Keep the Directions with the Most Variation7.517 min35 XPOpen
- 06t-SNE: A Neighbourhood Map, Not a Map of Everything7.615 min35 XPOpen
- 07Anomaly Detection: Unusual Does Not Automatically Mean Wrong7.716 min35 XPOpen
- 01Saving and Loading a Model: Keep What You Learned8.114 min30 XPOpen
- 02Pipelines: Keep Preprocessing and the Model Together8.216 min35 XPOpen
- 03Serving a Model as an API: Turn a Prediction into a Contract8.319 min40 XPOpen
- 04Deploying a Model: Release a Service You Can Roll Back8.416 min35 XPOpen
- 05After Launch: Monitoring and Drift8.516 min35 XPOpen
From your first ML concepts to shipping a model. Follow the path at your own pace.