Figure-Axes-Artist¶

Composite Pattern¶

1. Hierarchy¶

Figure └── Axes └── Artists (Line2D, Text, Patch, etc.)

2. Figure Object¶

Top-level container:

```python import matplotlib.pyplot as plt

fig = plt.figure(figsize=(10, 6)) print(type(fig)) # ```

3. Axes Object¶

Plotting region with coordinate system:

python fig, ax = plt.subplots() print(type(ax)) # <class 'matplotlib.axes._axes.Axes'>

Components¶

1. Figure Properties¶

python fig = plt.figure(figsize=(12, 8), dpi=100) fig.suptitle('Main Title') fig.tight_layout() fig.savefig('output.png')

2. Axes Properties¶

python fig, ax = plt.subplots() ax.set_xlabel('X Label') ax.set_ylabel('Y Label') ax.set_title('Axes Title') ax.set_xlim(0, 10) ax.set_ylim(0, 100) ax.grid(True)

3. Artist Objects¶

python line, = ax.plot([0, 1, 2], [0, 1, 0]) print(type(line)) # <class 'matplotlib.lines.Line2D'> line.set_color('red') line.set_linewidth(2)

Multiple Axes¶

1. Subplots¶

python fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 4)) ax1.plot([1, 2, 3]) ax2.scatter([1, 2, 3], [4, 5, 6])

2. GridSpec¶

```python from matplotlib.gridspec import GridSpec

fig = plt.figure(figsize=(10, 8)) gs = GridSpec(3, 3, figure=fig)

ax1 = fig.add_subplot(gs[0, :]) # Top row, all columns ax2 = fig.add_subplot(gs[1:, 0:2]) # Bottom 2 rows, first 2 cols ax3 = fig.add_subplot(gs[1:, 2]) # Bottom 2 rows, last col ```

3. Nested Axes¶

```python fig, ax = plt.subplots() ax.plot([1, 2, 3])

Inset axes¶

from mpl_toolkits.axes_grid1.inset_locator import inset_axes ax_inset = inset_axes(ax, width="30%", height="30%", loc='upper right') ax_inset.plot([1, 2, 3], [3, 2, 1]) ```

The Full Rendering Pipeline¶

The hierarchy doesn't stop at Artists. The complete model from data to screen:

text Data ↓ Artist (Line2D, Text, Patch, ...) ↓ Axes (coordinates, scaling, limits) ↓ Figure (container, size, DPI) ↓ Renderer + Backend (converts to pixels) ↓ Screen / File (.png, .pdf, .svg)

When something doesn't appear on screen, trace this pipeline: is the data correct? Is the Artist created? Are the axes limits right? Is the backend rendering?

Exercises¶

Exercise 1. Draw the Matplotlib object hierarchy from Figure down to Tick. Write code that accesses each level: fig, ax, ax.xaxis, ax.xaxis.get_major_ticks().

Exercise 2. Explain the parent-child relationship between Figure, Axes, Axis, and Tick objects.

Exercise 3. Write code that creates a figure with two subplots and demonstrates that both Axes share the same parent Figure using ax.get_figure().

Exercise 4. Access and modify a specific tick label on the x-axis programmatically. Change its color to red and increase its font size.

import matplotlib.pyplot as plt
import numpy as np

fig, ax = plt.subplots()
ax.plot(np.linspace(0, 10, 50), np.sin(np.linspace(0, 10, 50)))

# Draw first to populate tick labels
fig.canvas.draw()

# Access tick labels — they are Text Artists
labels = ax.get_xticklabels()
if len(labels) > 2:
    labels[2].set_color('red')
    labels[2].set_fontsize(16)

plt.show()

Exercise 5. Traverse the scene graph programmatically. Write code that creates a figure with 2 subplots, each containing a line plot. Then use fig.get_children() and ax.get_children() to print the full tree structure (Figure → Axes → Artists). Count the total number of Artists in the figure.

import matplotlib.pyplot as plt
import numpy as np

fig, (ax1, ax2) = plt.subplots(1, 2)
ax1.plot([1, 2, 3], [1, 4, 9], 'b-')
ax2.bar([1, 2, 3], [3, 1, 2])

# Traverse the scene graph
print(f"Figure children: {len(fig.get_children())}")
for i, child in enumerate(fig.get_children()):
    print(f"  [{i}] {type(child).__name__}")

print(f"\nAx1 children: {len(ax1.get_children())}")
for child in ax1.get_children()[:5]:
    print(f"  {type(child).__name__}")

# Count all Artists recursively
total = 0
for ax in fig.get_axes():
    total += len(ax.get_children())
print(f"\nTotal Artists across all Axes: {total}")

# This demonstrates the scene graph: Figure owns Axes,
# each Axes owns Lines, Patches, Text, Spines, etc.
plt.show()