Skip to main content

📊 Pandas & Data Analysis

Why Pandas? The Speed Problem​

If you did the Python 101 Hard track, you already timed your own pure-Python word-counting code on a text corpus — and felt it get sluggish as the corpus grew. That's not a coincidence: plain Python loops are slow for this kind of work. Pandas (built on numpy) exists precisely to fix that, by doing the looping in fast, vectorized C code under the hood instead of in Python itself. This section is about learning to use that speed.

🎯 Learning objectives — by the end of these 5 weeks you'll be able to:

  • Load, explore, and clean tabular data with pandas
  • Filter, group, and aggregate real datasets
  • Reproduce a Kaggle-style analysis end to end

🟢 Normal: Pandas Basics

Learn pandas fundamentals, then reproduce a classic Kaggle notebook end to end.

  • Series & DataFrame basics, reading CSV
  • Selection, filtering, indexing
  • Cleaning: missing values, dtypes, string ops
  • Groupby, aggregation, merging
  • Guided reproduction of a Titanic EDA notebook

⏱️ Time commitment: ~3–4 hours/week

Start Normal →

🔴 Hard: Full EDA Project

A project-based track: run a complete exploratory data analysis, framing questions, visualizing, and delivering a final report.

  • EDA framework: framing questions, profiling a dataset
  • Univariate analysis + visualizations
  • Bivariate/multivariate analysis, correlation
  • Advanced, storytelling visualizations
  • Final deliverable: a full EDA report on a real dataset

⏱️ Time commitment: ~4–6 hours/week

Start Hard →