Aspect Ratio Control

set_aspect() constrains the displayed image canvas so that one unit on the x-axis occupies the same number of pixels as one unit on the y-axis.

Three modes are shown:

1. Default — the image fills the panel exactly (no constraint).

2. set_aspect("equal") — forces a 1 : 1 pixel-per-unit ratio. A square array is displayed as a square image regardless of panel dimensions.

3. set_aspect(2.0) — makes the canvas twice as wide as it is tall.

The scale bar, axis ticks, and colorbar all update automatically when the canvas is resized by set_aspect.

import numpy as np
import anyplotlib as apl

rng = np.random.default_rng(0)

# ── Synthetic calibrated image ────────────────────────────────────────────────
N = 64
x = np.linspace(0, 10, N)   # nm
y = np.linspace(0, 10, N)
XX, YY = np.meshgrid(x, y)
data = (np.sin(XX) * np.cos(YY) + 0.2 * rng.standard_normal((N, N))).astype(np.float32)

# ── 1. Default (no aspect constraint) ────────────────────────────────────────

Default — no aspect constraint

The image stretches to fill the full panel area.

fig1, ax1 = apl.subplots(1, 1, figsize=(420, 340))
plot1 = ax1.imshow(data, axes=[x, y], units="nm", cmap="viridis")
plot1.set_title("Default (no constraint)")
plot1.set_xlabel("x (nm)")
plot1.set_ylabel("y (nm)")

fig1  # Interactive

# ── 2. Equal aspect ratio ─────────────────────────────────────────────────────

⚡

Equal aspect ratio

set_aspect("equal") (or equivalently set_aspect(1.0)) ensures that one nm on the x-axis occupies the same number of canvas pixels as one nm on the y-axis. The canvas height adjusts to be equal to the canvas width.

fig2, ax2 = apl.subplots(1, 1, figsize=(420, 340))
plot2 = ax2.imshow(data, axes=[x, y], units="nm", cmap="viridis")
plot2.set_aspect("equal")
plot2.set_title("set_aspect('equal')")
plot2.set_xlabel("x (nm)")
plot2.set_ylabel("y (nm)")

fig2  # Interactive

# ── 3. Explicit ratio ─────────────────────────────────────────────────────────

⚡

Explicit ratio (2 : 1)

set_aspect(2.0) makes the canvas twice as wide as it is tall. Useful when the physical x and y scales differ (e.g. a time-series scan where one axis is faster than the other).

fig3, ax3 = apl.subplots(1, 1, figsize=(420, 340))
plot3 = ax3.imshow(data, axes=[x, y], units="nm", cmap="inferno")
plot3.set_aspect(2.0)
plot3.set_title("set_aspect(2.0)")
plot3.set_xlabel("x (nm)")
plot3.set_ylabel("y (nm)")

fig3  # Interactive

⚡