Data Science and Machine Learning with Python

This repo is a collection of studies I'm doing in the field of Data Science and Machine Learning using Python. I'm going from the basics like statistics and linear algebra to more advanced topics.

📚 Knowledge Base Structure

This repository uses Obsidian for knowledge management. The structure is:

• notes/ - Obsidian vault with concepts, MOCs, and documentation
  • 00-inbox/ - Quick capture for new ideas
  • 01-concepts/ - Core concepts organized by topic (linear algebra, calculus, statistics, ML)
  • 02-literature/ - Papers, articles, and book notes
  • 03-projects/ - Project documentation
  • 04-MOCs/ - Maps of Content (hub notes linking related concepts)
  • templates/ - Note templates for consistent documentation
• code/ - Jupyter notebooks and Python scripts
  • notebooks/ - All Jupyter notebooks
  • scripts/ - Python modules and standalone scripts
  • experiments/ - Experimental code
• data/ - Datasets (not tracked in git by default)
  • raw/ - Original, immutable data
  • processed/ - Cleaned and transformed data
  • external/ - Data from external sources
• assets/ - Images, PDFs, and other resources

Using the Knowledge Base

1. Install Obsidian
2. Open this repository folder as a vault
3. Install recommended plugins:
   • Obsidian Git - Auto-backup to GitHub
   • Obsidian Wypst - Typst math rendering
   • Dataview - Query and organize notes
   • Templater - Note templates
   • Excalidraw - Drawing diagrams
   • Advanced Tables - Better table editing
4. Start with the MOCs in notes/04-MOCs/

Linking Code and Concepts

Concepts in notes/ link to implementations in code/notebooks/. Use the graph view to explore connections between theory and practice!

Setup Instructions

See obsidian-ml-spec.md for detailed installation and configuration instructions.

Topics Covered

• Linear Algebra
• Calculus
• Statistics & Probability
• Machine Learning Fundamentals
• Deep Learning
• Reinforcement Learning
• Data Visualization
• Feature Engineering

Getting Started

1. Clone this repository
2. Install dependencies (if any)
3. Open Obsidian and load this folder as a vault
4. Explore notebooks in code/notebooks/
5. Read concept notes in notes/01-concepts/