ML Prerequisites Refresher

The chapters in this module, in reading order.

#	Chapter
00	Classical ML Refresher - First-Principles Overview
01	Train-production gap - the model passed the wrong exam
02	Bias and variance — naming the two diseases
03	What shapes can your model actually draw?
04	Regularization — the soft leash that cures overthinking
05	Linear regression — drawing one line through the cloud
06	Gradient descent — the blindfolded hiker on a foggy mountain
07	Logistic regression — the simplest spam classifier + confidence score
08	Feature engineering — making the data linearly separable
09	Decision trees — branching boundaries from yes/no questions
10	Random forests — the voting panel, made literal
11	Gradient boosting and XGBoost — sequential trees correcting residuals
12	Evaluation and cross-validation — splits that mimic deployment
13	Metrics and calibration — picking and trusting numbers
14	Class imbalance and thresholds — when accuracy celebrates doing nothing
15	Support Vector Machines — the widest street between two document classes
16	K-Nearest Neighbors — ask the closest houses
17	Naive Bayes — update the odds, one clue at a time
18	PCA and Dimensionality Reduction — rotate the data, keep the signal
19	K-Means, DBSCAN, and Clustering — find groups before anyone names them
20	Honest admission — what classical ML cannot do well