🎯 catplot

base plot

# Kind Options: "point", "bar", "strip", "swarm", "box", "violin", "boxen"
 
# Basic boxplot by gender
sns.catplot(x='gender', y='math score', data=df, kind='box')
 
# Add row = lunch (standard vs free/reduced)
sns.catplot(x='gender', y='math score', data=df, kind='box', row='lunch')
 
# Add column = test prep course (completed vs none)
sns.catplot(x='gender', y='math score', data=df, kind='box', row='lunch', col='test preparation course')

📝 Use Case: Ideal when comparing how different categories (like gender, lunch, test prep) impact scores.

---

🔁 PairGrid

A customizable version of pairplot for examining relationships between all numerical features

# Simple layout: upper = scatter, diag = kde, lower = kde (in red)
g = sns.PairGrid(df)
g = g.map_upper(sns.scatterplot)
g = g.map_diag(sns.kdeplot, lw=2)
g = g.map_lower(sns.kdeplot, color="red")
 

📌 Note: If you see a warning about marker in kde plots, it’s safe to ignore — kde doesn’t use markers.

---

🗂️ FacetGrid

Create a grid of plots conditioned on categorical variables

# Create grid layout: columns by gender, rows by lunch
g = sns.FacetGrid(data=df, col='gender', row='lunch')
 
# Map a scatter plot to each subplot
g = g.map(plt.scatter, "math score", "reading score", edgecolor="w")
 
# Add legend
g.add_legend()
 
# Adjust spacing between plots
plt.subplots_adjust(hspace=0.4, wspace=1)

🔍 Use Case: Perfect for comparing bivariate relationships (like math vs reading scores) across different subgroups.

🧩 Why use Grid Plots?

Because they help you compare things side by side.

---

✅ Imagine this:

You want to see:

• Do boys and girls score differently?
• Does lunch type affect scores?
• What happens if they did a prep course?

One plot can’t show all of that.
But grid plots split the data into small groups and show many small plots — so you can compare quickly and clearly.

---

🧪 What each grid does:

Type	Use for…	Looks like
catplot	Compare groups	Boxplots per gender/lunch/prep
FacetGrid	Show many scatter plots split by category	One plot per group
PairGrid	Show how numbers relate (math vs reading, etc.)	A big grid of scatter and curves

---

🥪 Analogy:

Grid plots are like a sandwich tray.

Each small sandwich (plot) shows part of the full picture.
Instead of one big messy sandwich, you get bite-sized, easy-to-read plots.