All four title positions at once

A 2x2 arrangement of four 2x2 quadrants, each its own group – built with GroupLayout instead of hand-slicing a flat 4x4 grid (axes[0:2, 0:2] and friends): describe each quadrant’s own shape once, and subplots_from_groups() works out the combined grid and every axes’ real position itself. Each quadrant uses the one title_position that actually faces an outer edge from where it sits: top-left (“top”), top-right (“right”), bottom-left (“left”), bottom-right (“bottom”). All four reach a true edge of the figure, so tight_layout() reserves outer margin for every one of them automatically – including a left/right title’s own rendered width, not just a top/bottom title’s height.

That reservation only ever applies to the figure’s own outer edges, though: the boundary between the top two quadrants and the bottom two is an interior row gap, and the one between the left two and right two is an interior column gap – neither is faced by any of the four groups’ titles, so nothing widens either one for them automatically. Left at their defaults, the top pair’s box and the bottom pair’s box would collide at the row seam, and likewise left/right at the column seam. group_spacing(wspace=..., hspace=...) gives every box room on every side, interior boundaries included, without discarding the outer-edge margins tight_layout() already reserved for all four titles above.

Finding a quadrant again afterward by its outer position is one piece of a larger lookup API – see Finding a group or axes again: title, id, or position for the rest of it (by title, by id, many=True) in one place.

This figure has 16 panels — too many to usefully embed at a fixed size.

Open the full interactive example in a new page to pick a tool, then zoom, pan, point-pick or annotate any panel.

View this figure’s Vega export ↗ — the raw JSON spec, rendered live by a real Vega engine.

View this figure’s Vega-Lite export ↗ — the raw JSON spec(s), rendered live by a real Vega-Lite engine.

bottom-right quadrant's own title: Bottom-right

import numpy as np
import plotpress

rng = np.random.default_rng(9)
x = np.linspace(0, 6, 13)
y = np.linspace(0, 6, 13)
X, Y = np.meshgrid(x, y)

layout = plotpress.GroupLayout(2, 2)
layout.add_group(0, 0, 2, 2, title="Top-left", title_position="top", color="#d62728")
layout.add_group(0, 1, 2, 2, title="Top-right", title_position="right", color="#1f77b4")
layout.add_group(1, 0, 2, 2, title="Bottom-left", title_position="left", color="#2ca02c")
layout.add_group(1, 1, 2, 2, title="Bottom-right", title_position="bottom", color="#9467bd")

fig, axes = plotpress.subplots_from_groups(layout, figsize=(10, 9))
for outer_r in range(2):
    for outer_c in range(2):
        for r in range(2):
            for c in range(2):
                ax = axes[outer_r, outer_c][r, c]
                Z = (np.sin(X * 0.6 + r) * np.cos(Y * 0.6 + c)
                     + 0.05 * rng.standard_normal(X.shape))
                ax.pcolormesh(x, y, Z, cmap="cividis", vmin=-1.2, vmax=1.2)
                ax.tick_params(labelsize=6)

fig.group_spacing(wspace=40.0, hspace=40.0)
fig.tight_layout()

# Find a group by its outer (row, col) in the layout -- not by remembering
# which of the four it happened to be titled, or hand-slicing axes[2:4,
# 2:4] the way it would look without GroupLayout at all.
bottom_right = fig.get_group(row=1, col=1)
print("bottom-right quadrant's own title:", bottom_right.title)