Advanced matplotlib: Complete API reference & best practices

A comprehensive guide to matplotlib's advanced features, focusing on precise control, professional styling, and production-ready visualizations.

1. Configuration & environment setup

1.1. Global configuration with rcParams

import matplotlib.pyplot as plt
import matplotlib as mpl
import numpy as np
import warnings

# Suppress font warnings
warnings.filterwarnings('ignore', message='findfont')

# Complete rcParams configuration
mpl.rcParams.update({
    # Figure properties
    'figure.figsize': (10, 6),
    'figure.dpi': 100,
    'figure.facecolor': 'white',
    'figure.edgecolor': 'none',
    'figure.autolayout': False,
    'figure.max_open_warning': 20,

    # Axes properties
    'axes.facecolor': 'white',
    'axes.edgecolor': '#333333',
    'axes.linewidth': 1.0,
    'axes.grid': True,
    'axes.titlesize': 14,
    'axes.titleweight': 'bold',
    'axes.labelsize': 12,
    'axes.labelweight': 'normal',
    'axes.labelcolor': '#333333',
    'axes.axisbelow': True,
    'axes.prop_cycle': plt.cycler('color', 
        ['#1f77b4', '#ff7f0e', '#2ca02c', '#d62728', '#9467bd',
         '#8c564b', '#e377c2', '#7f7f7f', '#bcbd22', '#17becf']),

    # Grid properties
    'grid.color': '#b0b0b0',
    'grid.linestyle': '--',
    'grid.linewidth': 0.5,
    'grid.alpha': 0.5,

    # Font properties
    'font.family': 'sans-serif',
    'font.sans-serif': ['DejaVu Sans', 'Arial', 'Helvetica'],
    'font.size': 10,

    # Tick properties
    'xtick.labelsize': 10,
    'ytick.labelsize': 10,
    'xtick.color': '#333333',
    'ytick.color': '#333333',
    'xtick.direction': 'out',
    'ytick.direction': 'out',

    # Legend properties
    'legend.frameon': True,
    'legend.framealpha': 0.9,
    'legend.facecolor': 'white',
    'legend.edgecolor': '#CCCCCC',
    'legend.fontsize': 10,

    # Save figure properties
    'savefig.dpi': 300,
    'savefig.bbox': 'tight',
    'savefig.pad_inches': 0.1,
    'savefig.facecolor': 'white',
    'savefig.edgecolor': 'none'
})

# Demonstration plot
x = np.linspace(0, 10, 100)
fig, ax = plt.subplots()
for i in range(5):
    ax.plot(x, np.sin(x + i*0.5), label=f'Wave {i+1}')

ax.set_title('Default Configuration Applied')
ax.set_xlabel('X axis')
ax.set_ylabel('Y axis')
ax.legend(loc='upper right')
plt.show()

print("✅ Global configuration applied successfully")

1.2. Context managers for temporary settings

import matplotlib.pyplot as plt
import numpy as np

x = np.linspace(0, 10, 100)
y = np.sin(x)

# Create subplots to show the difference
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 5))

# Normal plot
ax1.plot(x, y)
ax1.set_title('Default Style')
ax1.grid(True)

# Using context manager for temporary style
with plt.rc_context({'lines.linewidth': 3, 
                      'lines.linestyle': '--',
                      'axes.facecolor': '#f0f0f0',
                      'grid.alpha': 0.8}):
    ax2.plot(x, y, color='red')
    ax2.set_title('Temporary Style with Context Manager')
    ax2.grid(True)

plt.tight_layout()
plt.show()

# Style contexts can also be used with 'with' statements
with plt.style.context('seaborn-v0_8-darkgrid'):
    fig, ax = plt.subplots(figsize=(8, 4))
    ax.plot(x, np.cos(x), 'g-', linewidth=2)
    ax.set_title('Using Seaborn Dark Grid Style Context')
    plt.show()

2. Advanced figure and axes management

2.1. GridSpec for complex layouts

import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec
import numpy as np

# Create figure
fig = plt.figure(figsize=(12, 8))

# Create GridSpec with custom spacing
gs = gridspec.GridSpec(3, 3, 
                       width_ratios=[1, 2, 1],   # Column width ratios
                       height_ratios=[1, 2, 1],  # Row height ratios
                       wspace=0.3,               # Width space between subplots
                       hspace=0.3)               # Height space between subplots

# Create subplots using GridSpec indices
ax1 = fig.add_subplot(gs[0, :])      # Top row, all columns
ax2 = fig.add_subplot(gs[1, 0])      # Middle row, first column
ax3 = fig.add_subplot(gs[1, 1:])     # Middle row, columns 2-3
ax4 = fig.add_subplot(gs[2, :2])     # Bottom row, first two columns
ax5 = fig.add_subplot(gs[2, 2])      # Bottom row, last column

# Add sample data to each subplot
x = np.linspace(0, 10, 100)

ax1.plot(x, np.sin(x), 'b-')
ax1.set_title('Spanning Full Width')

ax2.bar(['A', 'B', 'C'], [3, 7, 5])
ax2.set_title('Bar Chart')

ax3.scatter(np.random.randn(50), np.random.randn(50), alpha=0.5)
ax3.set_title('Scatter Plot (2x width)')

ax4.plot(x, np.cos(x), 'r--')
ax4.set_title('Cosine Wave')

ax5.pie([30, 25, 20, 25], labels=['A', 'B', 'C', 'D'])
ax5.set_title('Pie')

fig.suptitle('Complex Layout with GridSpec', fontsize=16, y=0.98)
plt.show()

2.2. Nested and irregular layouts

import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec
import numpy as np

fig = plt.figure(figsize=(14, 8))

# Create outer grid
outer_grid = gridspec.GridSpec(2, 2, wspace=0.3, hspace=0.3)

# Top-left: single plot
ax1 = fig.add_subplot(outer_grid[0, 0])
ax1.plot(np.random.randn(100).cumsum())
ax1.set_title('Single Plot')

# Top-right: nested 2x2 grid
inner_grid_1 = gridspec.GridSpecFromSubplotSpec(
    2, 2, subplot_spec=outer_grid[0, 1], wspace=0.1, hspace=0.1)

for i in range(4):
    ax = fig.add_subplot(inner_grid_1[i // 2, i % 2])
    ax.plot(np.random.randn(50))
    ax.set_title(f'Nested {i+1}', fontsize=9)
    ax.tick_params(labelsize=8)

# Bottom-left: nested 3x1 grid
inner_grid_2 = gridspec.GridSpecFromSubplotSpec(
    3, 1, subplot_spec=outer_grid[1, 0], hspace=0.4)

for i in range(3):
    ax = fig.add_subplot(inner_grid_2[i, 0])
    ax.bar(['A', 'B', 'C'], np.random.randint(1, 10, 3))
    ax.set_title(f'Bar {i+1}', fontsize=10)

# Bottom-right: single large plot
ax_large = fig.add_subplot(outer_grid[1, 1])
x = np.linspace(0, 10, 100)
ax_large.plot(x, np.sin(x), 'b-', label='sin')
ax_large.plot(x, np.cos(x), 'r--', label='cos')
ax_large.set_title('Large Plot with Legend')
ax_large.legend()
ax_large.grid(True, alpha=0.3)

fig.suptitle('Nested GridSpec Layouts', fontsize=16, y=0.98)
plt.show()

3. Precise positioning and spacing control

3.1. Manual axes positioning

import matplotlib.pyplot as plt
import numpy as np

fig = plt.figure(figsize=(12, 8))

# Manual positioning: [left, bottom, width, height] in figure coordinates (0-1)
ax1 = fig.add_axes([0.1, 0.7, 0.35, 0.25])  # Top-left
ax2 = fig.add_axes([0.55, 0.7, 0.35, 0.25]) # Top-right
ax3 = fig.add_axes([0.1, 0.35, 0.35, 0.25]) # Middle-left
ax4 = fig.add_axes([0.55, 0.35, 0.35, 0.25]) # Middle-right
ax5 = fig.add_axes([0.1, 0.05, 0.8, 0.2])   # Bottom (wide)

# Inset axes
ax_inset = fig.add_axes([0.62, 0.77, 0.15, 0.1])  # Inside ax2

# Add data
x = np.linspace(0, 10, 100)

ax1.plot(x, np.sin(x), 'b-')
ax1.set_title('Sine Wave')

ax2.plot(x, np.cos(x), 'r-')
ax2.set_title('Cosine Wave')

ax3.scatter(np.random.randn(30), np.random.randn(30))
ax3.set_title('Scatter')

ax4.hist(np.random.randn(100), bins=20, alpha=0.7)
ax4.set_title('Histogram')

ax5.plot(x, x**2, 'g-', linewidth=2)
ax5.set_title('Wide Bottom Plot')
ax5.set_xlabel('X axis')

# Inset plot
ax_inset.plot(x[:20], np.cos(x[:20]), 'r-', linewidth=1)
ax_inset.set_title('Inset', fontsize=8)
ax_inset.tick_params(labelsize=6)
ax_inset.set_facecolor('#f0f0f0')

fig.suptitle('Manual Axes Positioning', fontsize=16, y=0.98)
plt.show()

# Print positions for reference
for i, ax in enumerate([ax1, ax2, ax3, ax4, ax5], 1):
    pos = ax.get_position()
    print(f"Ax{i}: left={pos.x0:.2f}, bottom={pos.y0:.2f}, width={pos.width:.2f}, height={pos.height:.2f}")

3.2. Fine-tuning with subplots_adjust

import matplotlib.pyplot as plt
import numpy as np
from pyodide.http import open_url
import pandas as pd

# Load GDP data
url = "https://raw.githubusercontent.com/datasets/gdp/master/data/gdp.csv"
df = pd.read_csv(open_url(url))

# Get data for specific countries
countries = ['United States', 'China', 'Japan', 'Germany']
latest_year = 2019

fig, axes = plt.subplots(2, 2, figsize=(14, 10))

for idx, (ax, country) in enumerate(zip(axes.flat, countries)):
    country_data = df[df['Country Name'] == country]
    country_data = country_data[country_data['Year'] >= 2000]

    ax.plot(country_data['Year'], country_data['Value']/1e12, 
            marker='o', markersize=4, linewidth=2)
    ax.set_title(f'{country} GDP Evolution', fontsize=12, pad=10)
    ax.set_xlabel('Year', fontsize=10)
    ax.set_ylabel('GDP (Trillion USD)', fontsize=10)
    ax.grid(True, alpha=0.3)
    ax.tick_params(labelsize=9)

# Fine-tune spacing
plt.subplots_adjust(
    left=0.08,    # Left margin
    right=0.95,   # Right margin
    top=0.92,     # Top margin
    bottom=0.08,  # Bottom margin
    wspace=0.25,  # Width space between subplots
    hspace=0.35   # Height space between subplots
)

fig.suptitle('GDP Comparison with Fine-tuned Spacing', fontsize=14, y=0.98)

# Add text annotation about spacing
fig.text(0.5, 0.02, 'Spacing: left=0.08, right=0.95, top=0.92, bottom=0.08, wspace=0.25, hspace=0.35',
         ha='center', fontsize=9, style='italic', color='gray')

plt.show()

4. Advanced annotation and text control

4.1. Comprehensive annotation options

import matplotlib.pyplot as plt
import matplotlib.patches as patches
import numpy as np

fig, ax = plt.subplots(figsize=(12, 8))

# Sample data
x = np.linspace(0, 10, 100)
y = np.sin(x) * np.exp(-x/10)

ax.plot(x, y, 'b-', linewidth=2, label='Damped sine wave')

# Different annotation styles

# 1. Simple annotation with arrow
ax.annotate('Local maximum',
            xy=(1.5, 0.85), xytext=(3, 1.2),
            arrowprops=dict(arrowstyle='->', color='red', lw=1.5),
            fontsize=11, color='red')

# 2. Fancy arrow with bbox
ax.annotate('Important point',
            xy=(5, -0.1), xytext=(6.5, -0.4),
            arrowprops=dict(arrowstyle='fancy', 
                          connectionstyle='arc3,rad=0.3',
                          color='green', lw=2),
            bbox=dict(boxstyle='round,pad=0.5', 
                     facecolor='yellow', alpha=0.7),
            fontsize=10, fontweight='bold')

# 3. Curved arrow with custom connectionstyle
ax.annotate('Decay region',
            xy=(8, 0.05), xytext=(8.5, 0.3),
            arrowprops=dict(arrowstyle='wedge',
                          connectionstyle='arc3,rad=-0.3',
                          color='purple', lw=1.5),
            fontsize=10, color='purple')

# 4. Double-headed arrow
ax.annotate('', xy=(0.5, -0.5), xytext=(2.5, -0.5),
            arrowprops=dict(arrowstyle='<->', color='blue', lw=2))
ax.text(1.5, -0.6, 'Period', ha='center', fontsize=10, color='blue')

# 5. Text with rotation and alignment
ax.text(7, 0.6, 'Rotated text\nwith line break', 
        rotation=25, fontsize=11,
        ha='center', va='bottom',
        bbox=dict(boxstyle='sawtooth,pad=0.3', 
                 facecolor='lightblue', alpha=0.7))

# 6. Mathematical notation
ax.text(4, 0.7, r'$y = \sin(x) \cdot e^{-x/10}$',
        fontsize=14, color='black',
        bbox=dict(boxstyle='round,pad=0.5', 
                 facecolor='white', edgecolor='black'))

ax.set_xlim(0, 10)
ax.set_ylim(-0.8, 1.4)
ax.set_title('Advanced Annotation Techniques', fontsize=14, fontweight='bold')
ax.set_xlabel('X axis')
ax.set_ylabel('Y axis')
ax.grid(True, alpha=0.3)
ax.legend(loc='upper right')

plt.tight_layout()
plt.show()

5. Professional styling and themes

5.1. Custom style sheets

import matplotlib.pyplot as plt
import numpy as np

# Define custom style dictionary
custom_style = {
    # Colors
    'axes.prop_cycle': plt.cycler('color', 
        ['#E63946', '#F1FAEE', '#A8DADC', '#457B9D', '#1D3557']),

    # Figure
    'figure.facecolor': '#F8F9FA',
    'figure.edgecolor': 'none',

    # Axes
    'axes.facecolor': 'white',
    'axes.edgecolor': '#CCCCCC',
    'axes.linewidth': 1.5,
    'axes.grid': True,
    'axes.spines.top': False,
    'axes.spines.right': False,
    'axes.spines.left': True,
    'axes.spines.bottom': True,

    # Grid
    'grid.color': '#E0E0E0',
    'grid.linestyle': '-',
    'grid.linewidth': 0.8,
    'grid.alpha': 0.5,

    # Ticks
    'xtick.major.size': 5,
    'xtick.major.width': 1.5,
    'ytick.major.size': 5,
    'ytick.major.width': 1.5,
    'xtick.minor.visible': True,
    'ytick.minor.visible': True,
    'xtick.minor.size': 3,
    'ytick.minor.size': 3,

    # Font
    'font.size': 11,
    'axes.titlesize': 14,
    'axes.labelsize': 12,

    # Legend
    'legend.frameon': True,
    'legend.fancybox': True,
    'legend.shadow': True,
    'legend.framealpha': 0.9,
    'legend.edgecolor': '#CCCCCC'
}

# Apply custom style
with plt.rc_context(custom_style):
    fig, axes = plt.subplots(2, 2, figsize=(12, 8))

    x = np.linspace(0, 10, 100)

    # Different plot types with custom style
    axes[0, 0].plot(x, np.sin(x), linewidth=2, label='sin')
    axes[0, 0].plot(x, np.cos(x), linewidth=2, label='cos')
    axes[0, 0].set_title('Line Plot')
    axes[0, 0].legend()

    axes[0, 1].bar(['A', 'B', 'C', 'D'], [23, 45, 56, 78], alpha=0.8)
    axes[0, 1].set_title('Bar Chart')

    axes[1, 0].scatter(np.random.randn(50), np.random.randn(50), 
                      s=100, alpha=0.6, edgecolors='black', linewidth=1)
    axes[1, 0].set_title('Scatter Plot')

    axes[1, 1].hist(np.random.randn(100), bins=20, alpha=0.7, edgecolor='black')
    axes[1, 1].set_title('Histogram')

    fig.suptitle('Custom Professional Style', fontsize=16, fontweight='bold')
    plt.tight_layout()
    plt.show()

print("✅ Custom style applied successfully")

5.2. Publication-ready figures

import matplotlib.pyplot as plt
import numpy as np
from matplotlib.ticker import MultipleLocator, FormatStrFormatter

# Publication settings
plt.rcParams.update({
    'font.size': 10,
    'font.family': 'serif',
    'font.serif': ['Times New Roman', 'DejaVu Serif'],
    'text.usetex': False,  # Would be True for LaTeX rendering
    'axes.linewidth': 0.8,
    'lines.linewidth': 1.5,
    'lines.markersize': 6,
    'xtick.major.width': 0.8,
    'ytick.major.width': 0.8,
    'xtick.major.size': 4,
    'ytick.major.size': 4,
    'figure.dpi': 100,
    'savefig.dpi': 300
})

# Create figure with precise dimensions (in inches for publications)
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(7, 3.5))  # Common journal width

# Left panel: Multiple time series
x = np.linspace(0, 4*np.pi, 200)
y1 = np.sin(x) * np.exp(-x/10)
y2 = np.cos(x) * np.exp(-x/10)

ax1.plot(x, y1, 'k-', label='Dataset A', linewidth=1.5)
ax1.plot(x, y2, 'k--', label='Dataset B', linewidth=1.5)
ax1.fill_between(x, y1, y2, where=(y1 >= y2), alpha=0.3, color='gray', label='A > B')

ax1.set_xlabel('Time (s)', fontsize=11)
ax1.set_ylabel('Amplitude (a.u.)', fontsize=11)
ax1.set_title('(a) Time Series Comparison', fontsize=11, fontweight='bold', loc='left')

# Set precise tick locations
ax1.xaxis.set_major_locator(MultipleLocator(np.pi))
ax1.xaxis.set_minor_locator(MultipleLocator(np.pi/2))
ax1.xaxis.set_major_formatter(FormatStrFormatter('%.1f'))

ax1.legend(loc='upper right', fontsize=9, framealpha=1)
ax1.grid(True, alpha=0.3, which='major')
ax1.grid(True, alpha=0.1, which='minor')

# Right panel: Statistical plot
data = [np.random.normal(0, std, 100) for std in range(1, 4)]
positions = [1, 2, 3]

bp = ax2.boxplot(data, positions=positions, widths=0.6,
                 patch_artist=True,
                 boxprops=dict(facecolor='lightgray', color='black'),
                 whiskerprops=dict(color='black'),
                 capprops=dict(color='black'),
                 medianprops=dict(color='red', linewidth=2),
                 flierprops=dict(marker='o', markerfacecolor='gray', 
                                markersize=4, alpha=0.5))

ax2.set_xlabel('Group', fontsize=11)
ax2.set_ylabel('Value', fontsize=11)
ax2.set_title('(b) Statistical Distribution', fontsize=11, fontweight='bold', loc='left')
ax2.set_xticks(positions)
ax2.set_xticklabels(['Group 1', 'Group 2', 'Group 3'])
ax2.grid(True, alpha=0.3, axis='y')

# Add significance markers
y_max = 6
ax2.plot([1, 2], [y_max-0.5, y_max-0.5], 'k-', linewidth=1)
ax2.plot([1, 1], [y_max-0.5, y_max-0.7], 'k-', linewidth=1)
ax2.plot([2, 2], [y_max-0.5, y_max-0.7], 'k-', linewidth=1)
ax2.text(1.5, y_max-0.3, '**', ha='center', fontsize=12)

# Adjust layout with precise spacing
plt.tight_layout(pad=1.5)

# Add figure label
fig.text(0.02, 0.98, 'Figure 1', fontsize=12, fontweight='bold', 
         transform=fig.transFigure)

plt.show()

print("📊 Publication-ready figure created")
print("💡 For actual publication: set text.usetex=True for LaTeX rendering")

4.2. Text positioning and transforms

import matplotlib.pyplot as plt
import matplotlib.transforms as transforms
import numpy as np

fig, ax = plt.subplots(figsize=(10, 6))

# Sample data
x = np.linspace(0, 10, 50)
y = 2 * np.sin(x) + 1
ax.plot(x, y, 'b-', linewidth=2)

# Different coordinate systems

# 1. Data coordinates (default)
ax.text(5, 2, 'Data coords (5, 2)', 
        fontsize=10, color='blue',
        bbox=dict(boxstyle='round', facecolor='lightblue', alpha=0.5))

# 2. Axes coordinates (0-1 for both x and y)
ax.text(0.1, 0.9, 'Axes coords (0.1, 0.9)',
        transform=ax.transAxes, fontsize=10, color='red',
        bbox=dict(boxstyle='round', facecolor='pink', alpha=0.5))

# 3. Figure coordinates
fig.text(0.15, 0.02, 'Figure coords (0.15, 0.02)',
         fontsize=10, color='green',
         bbox=dict(boxstyle='round', facecolor='lightgreen', alpha=0.5))

# 4. Blended transforms (x in data, y in axes)
trans = transforms.blended_transform_factory(ax.transData, ax.transAxes)
ax.text(7, 0.5, 'Blended\n(x=data, y=axes)',
        transform=trans, fontsize=10, color='purple',
        ha='center', va='center',
        bbox=dict(boxstyle='round', facecolor='lavender', alpha=0.7))

# 5. Offset from data point
offset = transforms.ScaledTranslation(0.1, 0.1, fig.dpi_scale_trans)
shadow_transform = ax.transData + offset

# Plot point with shadow text
ax.plot(3, 1.5, 'ro', markersize=10)
ax.text(3, 1.5, 'Point', ha='center', va='center', color='white', fontweight='bold')
ax.text(3, 1.5, 'Shadow', transform=shadow_transform, 
        ha='center', va='center', color='gray', alpha=0.5)

ax.set_xlim(0, 10)
ax.set_ylim(-2, 4)
ax.set_title('Text Positioning with Different Coordinate Systems', fontsize=12, fontweight='bold')
ax.grid(True, alpha=0.3)

plt.tight_layout()
plt.show()

5. Professional styling and themes

5.1. Custom style sheets

import matplotlib.pyplot as plt
import numpy as np

# Define custom style dictionary
custom_style = {
    # Colors
    'axes.prop_cycle': plt.cycler('color', 
        ['#E63946', '#F1FAEE', '#A8DADC', '#457B9D', '#1D3557']),

    # Figure
    'figure.facecolor': '#F8F9FA',
    'figure.edgecolor': 'none',

    # Axes
    'axes.facecolor': 'white',
    'axes.edgecolor': '#CCCCCC',
    'axes.linewidth': 1.5,
    'axes.grid': True,
    'axes.spines.top': False,
    'axes.spines.right': False,
    'axes.spines.left': True,
    'axes.spines.bottom': True,

    # Grid
    'grid.color': '#E0E0E0',
    'grid.linestyle': '-',
    'grid.linewidth': 0.8,
    'grid.alpha': 0.5,

    # Ticks
    'xtick.major.size': 5,
    'xtick.major.width': 1.5,
    'ytick.major.size': 5,
    'ytick.major.width': 1.5,
    'xtick.minor.visible': True,
    'ytick.minor.visible': True,
    'xtick.minor.size': 3,
    'ytick.minor.size': 3,

    # Font
    'font.size': 11,
    'axes.titlesize': 14,
    'axes.labelsize': 12,

    # Legend
    'legend.frameon': True,
    'legend.fancybox': True,
    'legend.shadow': True,
    'legend.framealpha': 0.9,
    'legend.edgecolor': '#CCCCCC'
}

# Apply custom style
with plt.rc_context(custom_style):
    fig, axes = plt.subplots(2, 2, figsize=(12, 8))

    x = np.linspace(0, 10, 100)

    # Different plot types with custom style
    axes[0, 0].plot(x, np.sin(x), linewidth=2, label='sin')
    axes[0, 0].plot(x, np.cos(x), linewidth=2, label='cos')
    axes[0, 0].set_title('Line Plot')
    axes[0, 0].legend()

    axes[0, 1].bar(['A', 'B', 'C', 'D'], [23, 45, 56, 78], alpha=0.8)
    axes[0, 1].set_title('Bar Chart')

    axes[1, 0].scatter(np.random.randn(50), np.random.randn(50), 
                      s=100, alpha=0.6, edgecolors='black', linewidth=1)
    axes[1, 0].set_title('Scatter Plot')

    axes[1, 1].hist(np.random.randn(100), bins=20, alpha=0.7, edgecolor='black')
    axes[1, 1].set_title('Histogram')

    fig.suptitle('Custom Professional Style', fontsize=16, fontweight='bold')
plt.tight_layout()
plt.show()

print("✅ Custom style applied successfully")

5.2. Publication-ready figures

import matplotlib.pyplot as plt
import numpy as np
from matplotlib.ticker import MultipleLocator, FormatStrFormatter

# Publication settings
plt.rcParams.update({
    'font.size': 10,
    'font.family': 'serif',
    'font.serif': ['Times New Roman', 'DejaVu Serif'],
    'text.usetex': False,  # Would be True for LaTeX rendering
    'axes.linewidth': 0.8,
    'lines.linewidth': 1.5,
    'lines.markersize': 6,
    'xtick.major.width': 0.8,
    'ytick.major.width': 0.8,
    'xtick.major.size': 4,
    'ytick.major.size': 4,
    'figure.dpi': 100,
    'savefig.dpi': 300
})

# Create figure with precise dimensions (in inches for publications)
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(7, 3.5))  # Common journal width

# Left panel: Multiple time series
x = np.linspace(0, 4*np.pi, 200)
y1 = np.sin(x) * np.exp(-x/10)
y2 = np.cos(x) * np.exp(-x/10)

ax1.plot(x, y1, 'k-', label='Dataset A', linewidth=1.5)
ax1.plot(x, y2, 'k--', label='Dataset B', linewidth=1.5)
ax1.fill_between(x, y1, y2, where=(y1 >= y2), alpha=0.3, color='gray', label='A > B')

ax1.set_xlabel('Time (s)', fontsize=11)
ax1.set_ylabel('Amplitude (a.u.)', fontsize=11)
ax1.set_title('(a) Time Series Comparison', fontsize=11, fontweight='bold', loc='left')

# Set precise tick locations
ax1.xaxis.set_major_locator(MultipleLocator(np.pi))
ax1.xaxis.set_minor_locator(MultipleLocator(np.pi/2))
ax1.xaxis.set_major_formatter(FormatStrFormatter('%.1f'))

ax1.legend(loc='upper right', fontsize=9, framealpha=1)
ax1.grid(True, alpha=0.3, which='major')
ax1.grid(True, alpha=0.1, which='minor')

# Right panel: Statistical plot
data = [np.random.normal(0, std, 100) for std in range(1, 4)]
positions = [1, 2, 3]

bp = ax2.boxplot(data, positions=positions, widths=0.6,
                 patch_artist=True,
                 boxprops=dict(facecolor='lightgray', color='black'),
                 whiskerprops=dict(color='black'),
                 capprops=dict(color='black'),
                 medianprops=dict(color='red', linewidth=2),
                 flierprops=dict(marker='o', markerfacecolor='gray', 
                                markersize=4, alpha=0.5))

ax2.set_xlabel('Group', fontsize=11)
ax2.set_ylabel('Value', fontsize=11)
ax2.set_title('(b) Statistical Distribution', fontsize=11, fontweight='bold', loc='left')
ax2.set_xticks(positions)
ax2.set_xticklabels(['Group 1', 'Group 2', 'Group 3'])
ax2.grid(True, alpha=0.3, axis='y')

# Add significance markers
y_max = 6
ax2.plot([1, 2], [y_max-0.5, y_max-0.5], 'k-', linewidth=1)
ax2.plot([1, 1], [y_max-0.5, y_max-0.7], 'k-', linewidth=1)
ax2.plot([2, 2], [y_max-0.5, y_max-0.7], 'k-', linewidth=1)
ax2.text(1.5, y_max-0.3, '**', ha='center', fontsize=12)

# Adjust layout with precise spacing
plt.tight_layout(pad=1.5)

# Add figure label
fig.text(0.02, 0.98, 'Figure 1', fontsize=12, fontweight='bold', 
         transform=fig.transFigure)

plt.show()

print("📊 Publication-ready figure created")
print("💡 For actual publication: set text.usetex=True for LaTeX rendering")

6. Color management and colormaps

6.1. Advanced colormap usage

import matplotlib.pyplot as plt
import matplotlib.colors as mcolors
import numpy as np
from matplotlib.colors import LinearSegmentedColormap, ListedColormap

fig, axes = plt.subplots(2, 3, figsize=(14, 8))

# Generate sample data
x = np.linspace(-3, 3, 100)
y = np.linspace(-3, 3, 100)
X, Y = np.meshgrid(x, y)
Z = np.sin(np.sqrt(X**2 + Y**2))

# 1. Sequential colormap
im1 = axes[0, 0].imshow(Z, cmap='viridis', aspect='auto')
axes[0, 0].set_title('Sequential: Viridis')
plt.colorbar(im1, ax=axes[0, 0], fraction=0.046)

# 2. Diverging colormap with custom center
norm = mcolors.TwoSlopeNorm(vmin=Z.min(), vcenter=0, vmax=Z.max())
im2 = axes[0, 1].imshow(Z, cmap='RdBu_r', norm=norm, aspect='auto')
axes[0, 1].set_title('Diverging: RdBu (centered at 0)')
plt.colorbar(im2, ax=axes[0, 1], fraction=0.046)

# 3. Discrete colormap
n_bins = 8
cmap_discrete = plt.cm.get_cmap('plasma', n_bins)
im3 = axes[0, 2].imshow(Z, cmap=cmap_discrete, aspect='auto')
axes[0, 2].set_title(f'Discrete: Plasma ({n_bins} levels)')
plt.colorbar(im3, ax=axes[0, 2], fraction=0.046)

# 4. Custom colormap from list
colors_list = ['#FF0000', '#FF7F00', '#FFFF00', '#00FF00', '#0000FF']
n_bins = 100
cmap_custom = LinearSegmentedColormap.from_list('custom', colors_list, N=n_bins)
im4 = axes[1, 0].imshow(Z, cmap=cmap_custom, aspect='auto')
axes[1, 0].set_title('Custom: Rainbow from List')
plt.colorbar(im4, ax=axes[1, 0], fraction=0.046)

# 5. Logarithmic normalization
Z_positive = np.abs(Z) + 0.01  # Make all values positive
norm_log = mcolors.LogNorm(vmin=Z_positive.min(), vmax=Z_positive.max())
im5 = axes[1, 1].imshow(Z_positive, cmap='YlOrRd', norm=norm_log, aspect='auto')
axes[1, 1].set_title('Logarithmic Scale: YlOrRd')
plt.colorbar(im5, ax=axes[1, 1], fraction=0.046)

# 6. Bounded normalization
bounds = [-1, -0.5, 0, 0.5, 1]
norm_bounded = mcolors.BoundaryNorm(bounds, plt.cm.coolwarm.N)
im6 = axes[1, 2].contourf(X, Y, Z, levels=20, cmap='coolwarm', norm=norm_bounded)
axes[1, 2].set_title('Bounded Norm: Coolwarm')
plt.colorbar(im6, ax=axes[1, 2], fraction=0.046, boundaries=bounds, ticks=bounds)

fig.suptitle('Advanced Colormap Techniques', fontsize=14, fontweight='bold')
plt.tight_layout()
plt.show()

6.2. Color accessibility and best practices

import matplotlib.pyplot as plt
import numpy as np

# Colorblind-friendly palettes
colorblind_colors = {
    'blue': '#0173B2',
    'orange': '#DE8F05',
    'green': '#029E73',
    'red': '#CC78BC',
    'purple': '#ECE133',
    'brown': '#56B4E9',
    'pink': '#F0E442',
    'gray': '#999999'
}

fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Sample data
categories = list(colorblind_colors.keys())[:5]
values = [23, 45, 56, 78, 32]
x = np.linspace(0, 10, 100)

# Left: Bar chart with colorblind-friendly colors
colors = [colorblind_colors[cat] for cat in categories]
bars = ax1.bar(categories, values, color=colors, edgecolor='black', linewidth=1.5)

# Add patterns for additional distinction
patterns = ['/', '\\\\', '|', '-', '+']
for bar, pattern in zip(bars, patterns):
    bar.set_hatch(pattern)

ax1.set_title('Colorblind-Friendly with Patterns', fontsize=12, fontweight='bold')
ax1.set_ylabel('Values')
ax1.grid(axis='y', alpha=0.3)

# Right: Line plot with distinct styles
line_styles = ['-', '--', '-.', ':', '-']
markers = ['o', 's', '^', 'D', 'v']

for i, (cat, color) in enumerate(list(colorblind_colors.items())[:5]):
    y = np.sin(x + i) * np.exp(-x/10)
    ax2.plot(x, y, color=color, linestyle=line_styles[i], 
            marker=markers[i], markevery=10, markersize=8,
            linewidth=2, label=cat)

ax2.set_title('Multiple Line Styles for Accessibility', fontsize=12, fontweight='bold')
ax2.set_xlabel('X axis')
ax2.set_ylabel('Y axis')
ax2.legend(loc='upper right', framealpha=1)
ax2.grid(True, alpha=0.3)

fig.suptitle('Accessibility Best Practices', fontsize=14, fontweight='bold', y=1.02)

# Add note about accessibility
fig.text(0.5, -0.02, 
         '💡 Tip: Use both color and patterns/styles for maximum accessibility',
         ha='center', fontsize=10, style='italic', color='gray')

plt.tight_layout()
plt.show()

7. Performance optimization

7.1. Efficient plotting for large datasets

import matplotlib.pyplot as plt
import numpy as np
import time

# Generate large dataset
n_points = 100000
x = np.random.randn(n_points)
y = np.random.randn(n_points)

fig, (ax1, ax2, ax3) = plt.subplots(1, 3, figsize=(15, 5))

# Method 1: Regular scatter (slow for large data)
start = time.time()
ax1.scatter(x[:5000], y[:5000], alpha=0.5, s=1)
time1 = time.time() - start
ax1.set_title(f'Regular Scatter (5k points)\nTime: {time1:.3f}s')
ax1.set_xlabel('X')
ax1.set_ylabel('Y')

# Method 2: plot with markers (faster)
start = time.time()
ax2.plot(x[:50000], y[:50000], 'o', markersize=1, alpha=0.5)
time2 = time.time() - start
ax2.set_title(f'Plot with markers (50k points)\nTime: {time2:.3f}s')
ax2.set_xlabel('X')

# Method 3: hexbin for density (best for large data)
start = time.time()
hb = ax3.hexbin(x, y, gridsize=50, cmap='YlOrRd', mincnt=1)
time3 = time.time() - start
ax3.set_title(f'Hexbin (all {n_points} points)\nTime: {time3:.3f}s')
ax3.set_xlabel('X')
plt.colorbar(hb, ax=ax3, label='Count')

fig.suptitle('Performance Optimization for Large Datasets', fontsize=14, fontweight='bold')

plt.tight_layout()
plt.show()

print(f"\n📊 Performance Comparison:")
print(f"  Scatter (5k): {time1:.3f}s")
print(f"  Plot markers (50k): {time2:.3f}s")
print(f"  Hexbin (100k): {time3:.3f}s")
print(f"\n💡 Recommendation: Use hexbin or hist2d for datasets > 10k points")

7.2. Rasterization for complex plots

import matplotlib.pyplot as plt
import numpy as np

fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 5))

# Generate data
n = 10000
x = np.random.randn(n)
y = np.random.randn(n)

# Left: Vector graphics (large file size)
ax1.scatter(x, y, alpha=0.1, s=1, rasterized=False)
ax1.set_title('Vector Graphics\n(Large file, slow rendering)')
ax1.set_xlabel('X')
ax1.set_ylabel('Y')

# Right: Rasterized (small file size)
ax2.scatter(x, y, alpha=0.1, s=1, rasterized=True)
ax2.set_title('Rasterized\n(Small file, fast rendering)')
ax2.set_xlabel('X')
ax2.set_ylabel('Y')

# Add vector elements on top (not rasterized)
for ax in [ax1, ax2]:
    ax.axhline(0, color='red', linewidth=2, alpha=0.5)
    ax.axvline(0, color='red', linewidth=2, alpha=0.5)
    ax.grid(True, alpha=0.3)

fig.suptitle('Rasterization for Performance', fontsize=14, fontweight='bold')

plt.tight_layout()
plt.show()

print("💡 Rasterization benefits:")
print("  • Smaller file sizes for saved figures")
print("  • Faster rendering in viewers")
print("  • Text and axes remain vector (sharp)")
print("  • Use for: scatter plots, images, pcolormesh with many points")

8. Backend selection and LaTeX support

8.1. Understanding matplotlib backends

Matplotlib uses backends to render figures. The backend determines how plots are displayed and what features are available. When running locally (not in a browser-based environment), you can choose backends that support advanced features like LaTeX rendering.

Key backend types:

Interactive backends: Qt5Agg, TkAgg, MacOSX - for GUI windows
Non-interactive backends: Agg, PDF, PS, SVG - for file output only
Notebook backends: notebook, widget - for Jupyter environments

For more details, see the official matplotlib backend documentation.

8.2. LaTeX rendering with matplotlib

When running matplotlib locally, you can use LaTeX for beautiful mathematical typography. This requires a LaTeX installation on your system.

The latex below won't be displayed as latex when directly executed in a browser environment with pyodide. You need to run it locally.

import matplotlib.pyplot as plt
import numpy as np

# Note: LaTeX rendering requires local execution with LaTeX installed
# This example shows the syntax but won't render LaTeX in browser environments

# Configure matplotlib for LaTeX (for local execution)
plt.rcParams.update({
    'text.usetex': False,  # Set to True when running locally with LaTeX
    'font.family': 'serif',
    'font.serif': ['Computer Modern Roman'],
    'font.size': 12
})

# Create sample plot with mathematical notation
x = np.linspace(0, 2*np.pi, 100)
y1 = np.sin(x)
y2 = np.cos(x)

fig, ax = plt.subplots(figsize=(10, 6))

ax.plot(x, y1, 'b-', label=r'$\sin(x)$', linewidth=2)
ax.plot(x, y2, 'r--', label=r'$\cos(x)$', linewidth=2)

# Mathematical expressions in labels
ax.set_xlabel(r'$x$ (radians)', fontsize=14)
ax.set_ylabel(r'$f(x)$', fontsize=14)
ax.set_title(r'Trigonometric Functions: $\sin(x)$ and $\cos(x)$', fontsize=16)

# Add complex mathematical annotation
ax.text(np.pi, 0, r'$\int_{0}^{2\pi} \sin(x)\,dx = 0$',
        fontsize=12, ha='center', va='bottom',
        bbox=dict(boxstyle='round', facecolor='wheat', alpha=0.8))

# Another mathematical expression
ax.annotate(r'$\sin^2(x) + \cos^2(x) = 1$',
            xy=(np.pi/4, np.sin(np.pi/4)), xytext=(np.pi/4 + 1, 0.5),
            arrowprops=dict(arrowstyle='->', color='green', lw=1.5),
            fontsize=12, color='green')

ax.legend(loc='upper right', fontsize=12)
ax.grid(True, alpha=0.3)
ax.set_xlim(0, 2*np.pi)
ax.set_ylim(-1.5, 1.5)

plt.tight_layout()
plt.show()

print("📝 LaTeX rendering notes:")
print("• Requires LaTeX installation (TeX Live, MiKTeX, or MacTeX)")
print("• Set text.usetex=True in rcParams")
print("• Use raw strings (r'...') for LaTeX expressions")
print("• Expressions between $ signs for math mode")
print("• Full LaTeX commands available (\\frac, \\sqrt, \\int, etc.)")

8.3. Backend selection for local execution

python

# Example of backend selection (for local Python scripts)
import matplotlib
matplotlib.use('Qt5Agg')  # Set backend before importing pyplot
import matplotlib.pyplot as plt

# Or check current backend
print(matplotlib.get_backend())

# Common backend choices:
# - 'Qt5Agg': High-quality interactive backend (requires PyQt5)
# - 'TkAgg': Default interactive backend (uses Tkinter)
# - 'Agg': High-quality non-interactive backend
# - 'pdf': Direct PDF output
# - 'svg': Scalable vector graphics output

Installation for LaTeX support:

bash

# On macOS
brew install --cask mactex

# On Ubuntu/Debian
sudo apt-get install texlive-full

# On Windows
# Download and install MiKTeX from https://miktex.org/

# Python packages
pip install matplotlib[latex]

Summary and quick reference

Key takeaways

Configuration: Use rcParams for global settings, context managers for temporary changes
Layout: Master GridSpec for complex layouts, subplots_adjust for fine-tuning
Performance: Use appropriate plot types for data size, rasterize when needed
Styling: Create custom styles for consistency, follow accessibility guidelines
Export: Match format and DPI to intended use (web/print/publication)
Backends: Choose appropriate backend for your environment and needs

Essential commands reference

python

# Configuration
plt.rcParams.update({...})          # Global settings
with plt.rc_context({...}):         # Temporary settings

# Layout
gs = GridSpec(rows, cols, ...)      # Complex layouts
plt.subplots_adjust(...)            # Fine spacing control
fig.add_axes([l, b, w, h])         # Manual positioning

# Performance
ax.plot(..., rasterized=True)       # Rasterize dense plots
ax.hexbin(x, y, ...)                # Efficient for large data

# Styling
plt.style.use('style_name')         # Apply style
ax.set_prop_cycle(...)              # Custom color cycle

# Export
fig.savefig('file.pdf', dpi=300, bbox_inches='tight')
plt.close(fig)                      # Free memory

# Backend &amp; LaTeX
matplotlib.use('backend_name')      # Set backend
plt.rcParams['text.usetex'] = True  # Enable LaTeX

This comprehensive guide covers the advanced aspects of matplotlib not detailed in the basic tutorials, providing you with professional-level control over your visualizations.