Ch.02 Mini Project 03 — Distribution Visualizer¶

Concept: Univariate visualization — histograms, box plots, skewness, and common probability distributions (Normal and Uniform).

What it does¶

distribution_visualizer.py produces four types of plots for any numeric column:

Combined histogram + box plot (like the lecture slide p.45) — histogram on the bottom, box plot on top, sharing the x-axis
Histogram with Normal overlay — fits N(mu, sigma) to the data and overlays the PDF
Box plot with annotations — labels min, Q1, median, Q3, max
Normal vs. Uniform comparison — side-by-side PDF comparison to understand both distributions

Falls back to ASCII histograms if matplotlib is not installed.

Dependencies¶

pip install matplotlib

Usage¶

# Built-in demo (Ch.2 Friends dataset — Weight, Height, Max_temp)
python distribution_visualizer.py

# All numeric columns from a CSV
python distribution_visualizer.py my_data.csv

# Single column
python distribution_visualizer.py my_data.csv Age

What each plot shows¶

Combined histogram + box plot¶

Bin count chosen by the sqrt rule: bins ≈ √n
Rug plot (tick marks) below the histogram shows individual data points
Normal PDF overlay for visual distribution fit check
Skewness value and label in the title

Box plot anatomy (from lecture)¶

  whisker top ─── Max
                  |
  top of box  ─── Q3
  median line ─── Q2  (median)
  bottom of box── Q1
                  |
  whisker bottom─ Min

Normal vs. Uniform comparison¶

Normal N(mu, sigma): bell-shaped, symmetric, two parameters — mean (center) and std dev (width)
Uniform U(a, b): flat, equal probability everywhere in [a, b]

Key Concepts from Ch.2 Applied¶

Concept	Where it appears
Histogram bin selection (sqrt rule)	`suggest_bins()`
Normal PDF	`normal_pdf()`
Uniform PDF	`uniform_pdf()`
Skewness classification	`skewness()` + title annotation
Box plot anatomy	`plot_boxplot()`
Combined chart	`plot_combined()`

Limitations & Future Ideas¶

Extension: add Sturges' rule and Freedman-Diaconis rule as alternative bin selectors.
Extension: add a Q-Q plot to formally test normality.
Extension: add a kernel density estimate (KDE) as an alternative to the histogram.