How zotroneneis/machine_learning_basics works
Main Components Organization
a) Learning Algorithms Layer
• Algorithms into categories:
• Regression Models (Linear, Logistic, Bayesian, Softmax)
• Tree-based Models (Decision Trees for Classification and Regression)
• Neural Networks (Perceptron, Simple Neural Net)
• Clustering & Dimensionality Reduction (k-Means, PCA)
• Instance-based Learning (k-Nearest Neighbor)
• Support Vector Machines
b) Data Processing Layer
• Image Preprocessing Pipeline
• Numerical/Categorical Data Preprocessing Pipeline
c) Core Infrastructure Layer
• Jupyter Notebook Environment
• Python Runtime (3.6+)
• Dependencies (from requirements.txt)
Component Relationships
• Data flow from preprocessing components to learning algorithms
• Which algorithms can be used with which types of preprocessed data
• Infrastructure layer supporting both preprocessing and algorithms
[Data Sources] → [Preprocessing Layer] → [Algorithm Layer] → [Output/Predictions]
• ↑ ↑ ↑
• [Infrastructure Layer]──────────┘
Additional Elements
Annotations indicating:
• Implementation is from scratch (no external ML libraries)
• Educational purpose of each component
• Python version requirements
• Binder integration points