ML Foundations

beginner

What ML actually is, the data-to-model pipeline, the two workhorse algorithms (linear and logistic regression), decision trees, and how to honestly evaluate any of them. Complete.

What Is Machine Learning, Actually?

The ML Pipeline: From Raw Data to a Deployed Model

Linear Regression: Cost Functions and the Line of Best Fit

Logistic Regression: From Linear Scores to Probabilities

Decision Trees: Splits, Information Gain, and the ID3 Algorithm

Train/Test Splits & Cross-Validation: Why Random Splits Can Lie to You

ML Basics & Practical Notes

intermediate

How training actually adjusts a model's parameters, how to diagnose and fix over/underfitting, preparing data properly, combining models, and scoring them honestly. Complete.

Gradient Descent: Batch, Stochastic, and Mini-Batch

Overfitting vs Underfitting: Diagnosing Bias and Variance

Regularization: L1/L2, Early Stopping, and Dropout

Feature Engineering: Scaling, Normalization, and Encoding Categorical Data

Ensemble Methods: Random Forests, Bagging, and Gradient Boosting

Evaluation Metrics: Precision, Recall, F1, and Why Accuracy Lies

advanced

Learning without labels, catching outliers, explaining a model's decisions, and the classical algorithms this track set aside for last. Complete — all 20 pages.

reference

scikit-learn documentation
PyTorch documentation

ML Foundations

beginner

What Is Machine Learning, Actually?

The ML Pipeline: From Raw Data to a Deployed Model

Linear Regression: Cost Functions and the Line of Best Fit

Logistic Regression: From Linear Scores to Probabilities

Decision Trees: Splits, Information Gain, and the ID3 Algorithm

Train/Test Splits & Cross-Validation: Why Random Splits Can Lie to You

ML Basics & Practical Notes

intermediate

Gradient Descent: Batch, Stochastic, and Mini-Batch

Overfitting vs Underfitting: Diagnosing Bias and Variance

Regularization: L1/L2, Early Stopping, and Dropout

Feature Engineering: Scaling, Normalization, and Encoding Categorical Data

Ensemble Methods: Random Forests, Bagging, and Gradient Boosting

Evaluation Metrics: Precision, Recall, F1, and Why Accuracy Lies

advanced

Unsupervised Learning: K-Means Clustering and PCA

Anomaly Detection: Catching the Outliers Supervised Learning Misses

Data Leakage: The Silent Killer of Model Performance

Model Explainability: Feature Importance, Partial Dependence, and SHAP

Support Vector Machines and the Kernel Trick

Bayesian Learning and Naive Bayes

VC Dimension & Computational Learning Theory: Why Some Models Generalize

Instance-Based Learning: k-NN and the Curse of Dimensionality

related topics

reference

Contents

Get in touch