Home › Courses › Data Analysis & Visualization

Data Analysis & Visualization

pandas, NumPy, and the daily skill of turning real messy data into an answer

6 phases. 28 lessons. 28 labs. 1 capstone. The daily-practice depth this platform's other courses only touch in passing — hand-parsing painful data before pandas earns its keep, pandas mastery on pandas 3.0's current Copy-on-Write behavior, NumPy underneath it, matplotlib/seaborn visualization, and real exploratory data analysis. Built entirely on real, live-fetched datasets: actual Chipotle orders, NOAA weather stations, NYC taxi trips, and two real Inside Airbnb city snapshots.

6 phases · 28 lessons · 28 labs · 1 capstone · Level: Beginner

Take Programming & CS Foundations first — this course builds on it.

Outcomes you will have by the end

• A cleaned, analyzed, and reported real dataset — A real, messy 7,189-listing Airbnb snapshot explored, cleaned with documented reasoning, analyzed for genuine findings, and reported for a reader with zero context.
• Real judgment for messy real data — The habit of investigating an outlier against its own row's evidence, checking a finding across a different data slice, and reporting the real sample size behind every claim — not a mechanical cleaning checklist.
• Publication-quality chart skills — The object-oriented matplotlib pattern, seaborn built on top of it, and the specific labeling, color, and annotation choices that make a chart readable by a stranger.
• Verified certificate of completion — Issued by DeVenture Academy, tied to your completion record. Lists the pandas, NumPy, visualization, and EDA skills you built.

What you will be able to do

pandas · NumPy · Data Cleaning · Data Visualization · Exploratory Data Analysis · Statistical Reasoning

Every phase, every lesson, every project

• Working with Data in Python (5 lessons) — free — Hand-parsing real CSV data, hand-written aggregation bugs, meeting the DataFrame, fixed-width files, nested JSON.
  • Data Without a Library
  • Why This Gets Painful Fast
  • Meet the DataFrame
  • Reading Real, Ugly Files
  • JSON and Nested Data
• pandas Mastery (6 lessons) — free — Filtering and Copy-on-Write, sorting, groupby split-apply-combine, merging, missing data, time series — on real Chipotle, NYC taxi, and Airbnb data.
  • Selecting and Filtering
  • Sorting and Ranking
  • Split-Apply-Combine (groupby)
  • Joining and Merging
  • Missing Data, For Real
  • Time Series and Method Chaining
• NumPy for Data (4 lessons) — Fixed-type arrays, vectorized operations measured at a real 60x speedup, broadcasting, views vs. copies.
  • Arrays Aren't Just Fancy Lists
  • Vectorized Operations
  • Broadcasting
  • Indexing Tricks
• Data Visualization (5 lessons) — The object-oriented matplotlib pattern, combining seaborn with it, choosing a chart type from data shape, catching a misleading axis.
  • Figures and Axes, the Object-Oriented Way
  • Seaborn for Statistical Plots
  • Choosing the Right Chart
  • How Charts Lie
  • Publication-Quality Polish
• Exploratory Data Analysis (4 lessons) — EDA as question-generation, investigating real outliers individually, correlation vs. confounding variables, checking findings across data slices.
  • What EDA Actually Is
  • Outliers and Data Quality
  • Correlation Is Not the Whole Story
  • Asking Good Questions of Data
• Capstone (4 lessons) — A real, never-seen 7,189-listing New Orleans Airbnb snapshot — explored, cleaned with documented reasoning, analyzed, and reported.
  • Choose and Load
  • Clean
  • Analyze and Visualize
  • Report

The technologies you will use

pandas · NumPy · matplotlib · seaborn

Roles this course prepares you for

• Data Analyst ($65k–$90k) — Clean, analyze, and visualize real business data to answer specific questions — the exact daily loop this course builds, on genuinely messy real datasets, not toy CSVs.
• Junior Data Scientist / ML Engineer ($85k–$130k) — Prepare and explore data before any model gets trained on it — the unglamorous majority of real ML work, and the specific gap this course fills before Core ML from Scratch.

Common questions

Is this pandas content current, or from an older tutorial?

Current — every example is verified against real pandas 3.0.5, including Copy-on-Write behavior that changed how chained assignment and inplace=True actually work. Many older tutorials teach patterns that no longer behave the way they describe.

Are the datasets real, or simplified for teaching?

Real — this course uses the actual public Chipotle orders dataset, real NOAA weather station data, a full real month of NYC taxi trip records, and two real, current Inside Airbnb city snapshots. No iris, no titanic.

Do I need Programming & CS Foundations first?

Yes, or equivalent real Python fluency — this course starts with hand-parsing real data using core Python before pandas is introduced, specifically so meeting the DataFrame feels like a relief, not an arbitrary new API.

How is this different from the data content in ML & AI Engineering?

That course covers pandas and NumPy in a couple of setup lessons — enough to use the libraries. This course spends 28 lessons on the daily-practice depth: real data-quality judgment calls, EDA methodology, and chart-selection reasoning most ML courses only touch in passing.

Continue your learning path

Programming & CS Foundations · Math for AI & ML · ML & AI Engineering · Core ML: Algorithms from Scratch

Start the Data Analysis & Visualization course

Create a free account — the opening phases of 24 of 30 courses are free, no credit card. Or see Pro pricing.

All courses · Pricing · About · FAQ · Glossary