Shelf collection

Data science and machine learning notes.

Machine learning chapter notes, model concepts, and practical lessons synced from my learning vault.

Machine Learning

9 notes

The Machine Learning Landscape

A broad overview of machine learning systems, tasks, and the workflow for turning data into useful models.

Machine learning workflow from source data to model performance

End-to-End Machine Learning

Practical project workflow notes covering data preparation, validation, model training, and evaluation.

Classification

Classification concepts, metrics, decision thresholds, and ways to evaluate model performance.

Model training workflow from data through loss optimization to a trained model

Training Models

Core training ideas including regression, gradient descent, regularization, and optimization behavior.

Support vector machine diagram showing the maximum margin and optimal hyperplane

Support Vector Machines

Support vector machines, margins, kernels, similarity features, and related optimization ideas.

Decision tree diagram branching into two classification outcomes

Decision Trees

Decision tree intuition, splitting criteria, complexity, overfitting, and random forest foundations.

Random forest ensemble combining decisions from multiple trees into one prediction

Ensemble Learning and Random Forests

Bagging, boosting, random forests, and ensemble methods for improving model performance.

High-dimensional data projected into a lower-dimensional representation

Dimensionality Reduction

Dimensionality reduction, principal component analysis, manifold learning, and techniques for preserving useful structure in lower-dimensional spaces.

Unlabeled data organized into discovered clusters through unsupervised learning

Unsupervised Learning Techniques

Clustering, anomaly detection, density estimation, K-means, DBSCAN, and other techniques for learning patterns from unlabeled data.