AI From Scratch
Phase 02/18 lessons/~21 hours

ML Fundamentals

Classical machine learning — still the backbone of most production AI.

0 / 18 complete0%
Start phase
Lessons
01What Is Machine LearningUp nextMachine learning is teaching computers to find patterns in data instead of writing rules by hand.Learn/~45 minutes/Python02Linear RegressionLinear regression draws the best straight line through your data. It is the "hello world" of machine learning.Build/~90 minutes/Python03Logistic RegressionLogistic regression bends a straight line into an S-curve to answer yes-or-no questions with probabilities.Build/~90 minutes/Python04Decision Trees and Random ForestsA decision tree is just a flowchart. But a forest of them is one of the most powerful tools in ML.Build/~90 minutes05Support Vector MachinesFind the widest street between two classes. That is the entire idea.Build/~90 minutes06K-Nearest Neighbors and DistancesStore everything. Predict by looking at your neighbors. The simplest algorithm that actually works.Build/~90 minutes07Unsupervised LearningNo labels, no teacher. The algorithm finds structure on its own.Build/~90 minutes/Python08Feature Engineering & SelectionA good feature is worth a thousand data points.Build/~90 minutes/Python09Model EvaluationA model is only as good as the way you measure it.Build/~90 minutes/Python10Bias-Variance TradeoffEvery model error comes from one of three sources: bias, variance, or noise. You can only control the first two.Learn/~75 minutes11Ensemble MethodsA group of weak learners, combined correctly, becomes a strong learner. This is not a metaphor. It is a theorem.Build/~120 minutes12Hyperparameter TuningHyperparameters are the knobs you turn before training starts. Turning them well is the difference between a mediocre model and a great one.Build/~90 minutes13ML PipelinesA model is not a product. A pipeline is. The pipeline is everything from raw data to deployed prediction, and every step must be reproducible.Build/~120 minutes14Naive BayesThe "naive" assumption is wrong, and it works anyway. That's the beauty of it.Build/~75 minutes15Time Series FundamentalsPast performance does predict future results -- if you check for stationarity first.Build/~90 minutes16Anomaly DetectionNormal is easy to define. Abnormal is whatever doesn't fit.Build/~75 minutes17Handling Imbalanced DataWhen 99% of your data is "normal," accuracy is a lie.Build/~90 minutes18Feature SelectionMore features is not better. The right features is better.Build/~75 minutes