Course Wiki

Pandas

The DataFrame toolkit every SMA Zoomcamp module runs on: loading, cleaning, joining, and reshaping market data.

Module 2 is the data-handling core: NumPy, pandas, and the visualization stack (Matplotlib, Seaborn, Plotly Express) applied to financial data. The work covers numeric, string, and date/time types; cleaning strategies for messy market data; joining multiple tickers’ datasets into one analysis frame; and descriptive statistics with correlations across stocks.

Feature generation is the module’s output skill: deriving growth over day/week/month windows, calendar features like day-of-week, and the numeric foundations the modeling module consumes. Everything is Colab-first so the analysis runs anywhere.

Taught in

• Machine Learning Zoomcamp

  • Module 1: Introduction to Machine Learning
    • Setting up the Environment
    • Introduction to Pandas
    • Summary
  • Module 2: Machine Learning for Regression
    • Data preparation
    • Exploratory data analysis
    • Setting up the validation framework
    • Categorical variables
  • Module 3: Machine Learning for Classification
    • Data preparation
    • Setting up the validation framework
    • EDA
    • Feature importance: Churn rate and risk ratio
    • Feature importance: Mutual information
    • Feature importance: Correlation
  • Module 6: Decision Trees and Ensemble Learning
    • Data cleaning and preparation
    • Decision trees parameter tuning
    • Gradient boosting and XGBoost
    • XGBoost parameter tuning

• LLM Zoomcamp

  • Module 2: Vector Search
    • Vector Search
  • Module 4: Evaluation
    • Generating Ground Truth for All Documents
    • Search Evaluation
    • Search Parameter Tuning
    • Generating RAG Answers
    • LLM as a Judge
    • Agent Evaluation
  • Module 5: Monitoring
    • Streamlit Dashboard
  • Module 7: End-to-End Project
    • End-to-End Project Example
    • Evaluating Retrieval
    • Evaluating RAG
    • Interface and Ingestion Pipeline

• Stock Markets Analytics Zoomcamp

  • Module 2: Working with the Data
    • Working with the Data ## Related concepts

• Market Data APIs

• Technical Indicators

• Time Series Decomposition

• Backtesting