Seaborn Multi-Plot Grids

🔲 pairplot — every variable vs every other

sns.pairplot(df)                                  # all numeric columns, scatter + histogram on diagonal
sns.pairplot(df, hue="species")                       # color by category
sns.pairplot(df, vars=["col1", "col2", "col3"])          # limit to specific columns
sns.pairplot(df, diag_kind="kde")                           # KDE instead of histogram on the diagonal
sns.pairplot(df, kind="reg")                                   # regression line in off-diagonal cells
sns.pairplot(df, corner=True)                                      # only lower triangle (skip redundant mirror)

pairplot is usually the fastest first step in exploratory analysis — one call reveals every pairwise relationship + each variable's distribution at once.

🎯 jointplot — two variables + their individual distributions

sns.jointplot(data=df, x="total_bill", y="tip")
sns.jointplot(data=df, x="total_bill", y="tip", kind="scatter")      # default
sns.jointplot(data=df, x="total_bill", y="tip", kind="hex")             # hexbin, good for dense data
sns.jointplot(data=df, x="total_bill", y="tip", kind="kde")                # 2D density
sns.jointplot(data=df, x="total_bill", y="tip", kind="reg")                   # scatter + regression line
sns.jointplot(data=df, x="total_bill", y="tip", hue="time")                      # color by category

Shows a central scatter/density plot + marginal histograms on the top and right edges.

🧩 FacetGrid — manual, flexible faceting

g = sns.FacetGrid(df, col="time", row="sex", hue="smoker")
g.map(sns.scatterplot, "total_bill", "tip")
g.add_legend()

g = sns.FacetGrid(df, col="day", col_wrap=2, height=3)
g.map_dataframe(sns.histplot, x="total_bill")      # map_dataframe passes the actual sub-DataFrame, not raw arrays

Reach for FacetGrid directly when a figure-level function (relplot/catplot/displot) doesn't support the exact plot type you need — those are actually built on top of FacetGrid internally.

Method	Passes to the plotting function
.map(func, "x", "y")	positional arrays
.map_dataframe(func, x="x", y="y")	keyword args + the facet’s own sub-DataFrame

🔗 PairGrid — manual version of pairplot

g = sns.PairGrid(df, hue="species")
g.map_diag(sns.histplot)
g.map_offdiag(sns.scatterplot)
g.add_legend()

g = sns.PairGrid(df)
g.map_upper(sns.scatterplot)
g.map_lower(sns.kdeplot)
g.map_diag(sns.histplot)

PairGrid lets you use a DIFFERENT plot type for upper triangle, lower triangle, and diagonal — pairplot can't do that mix in one call.

🎯 JointGrid — manual version of jointplot

g = sns.JointGrid(data=df, x="total_bill", y="tip")
g.plot(sns.scatterplot, sns.histplot)

🔗 Next

Seaborn Styling & Themes · Seaborn Statistical Estimation & Data Handling