matplotlib-best-practices

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Matplotlib Best Practices

Matplotlib 最佳实践

Expert guidelines for Matplotlib development, focusing on data visualization, plotting, and creating publication-quality figures.

Matplotlib开发的专家指南，聚焦于数据可视化、绘图及生成出版级图表。

Code Style and Structure

代码风格与结构

Write concise, technical Python code with accurate Matplotlib examples
Create informative and visually appealing plots with proper labels, titles, and legends
Use the object-oriented API for complex figures, pyplot for quick plots
Follow PEP 8 style guidelines
Consider color-blindness accessibility in all visualizations

编写简洁、专业的Python代码，搭配准确的Matplotlib示例
创建带有恰当标签、标题和图例的信息丰富且视觉美观的图表
针对复杂图表使用面向对象API，快速绘图使用pyplot
遵循PEP 8风格指南
在所有可视化中考虑色盲友好的可访问性

API Approaches

API 方法

Object-Oriented Interface (Recommended)

面向对象接口（推荐）

Use
```
fig, ax = plt.subplots()
```
for explicit control
Preferred for complex figures and production code
Methods are called on axes objects:
```
ax.plot()
```
,
```
ax.set_xlabel()
```
Enables multiple subplots and fine-grained customization

使用
```
fig, ax = plt.subplots()
```
实现显式控制
适用于复杂图表和生产代码
方法调用在轴对象上：
```
ax.plot()
```
、
```
ax.set_xlabel()
```
支持多子图和精细化定制

Pyplot Interface

Pyplot 接口

Use
```
plt.plot()
```
,
```
plt.xlabel()
```
for quick, interactive plots
Suitable for Jupyter notebooks and exploration
Use
```
%matplotlib inline
```
in Jupyter notebooks

使用
```
plt.plot()
```
、
```
plt.xlabel()
```
进行快速交互式绘图
适用于Jupyter笔记本和探索性分析
在Jupyter笔记本中使用
```
%matplotlib inline
```

Creating Effective Visualizations

创建高效可视化

Plot Types and Selection

图表类型选择

Line plots (
```
ax.plot()
```
) for continuous data and trends
Scatter plots (
```
ax.scatter()
```
) for relationship between variables
Bar plots (
```
ax.bar()
```
,
```
ax.barh()
```
) for categorical comparisons
Histograms (
```
ax.hist()
```
) for distributions
Box plots (
```
ax.boxplot()
```
) for statistical summaries
Heatmaps (
```
ax.imshow()
```
,
```
ax.pcolormesh()
```
) for 2D data

折线图（
```
ax.plot()
```
）用于展示连续数据和趋势
散点图（
```
ax.scatter()
```
）用于展示变量间的关系
柱状图（
```
ax.bar()
```
、
```
ax.barh()
```
）用于分类比较
直方图（
```
ax.hist()
```
）用于展示分布情况
箱线图（
```
ax.boxplot()
```
）用于统计摘要
热力图（
```
ax.imshow()
```
、
```
ax.pcolormesh()
```
）用于二维数据

Labels and Annotations

标签与注释

Always include axis labels with units
Use descriptive titles that convey the message
Add legends when multiple series are present
Use annotations (
```
ax.annotate()
```
) to highlight key points
Include data source attribution when appropriate

始终包含带单位的轴标签
使用能传达核心信息的描述性标题
当存在多个系列时添加图例
使用注释（
```
ax.annotate()
```
）突出关键点
适当情况下注明数据来源

Color and Style

颜色与样式

Use colorblind-friendly palettes (e.g., 'viridis', 'plasma', 'cividis')
Avoid red-green combinations for accessibility
Use consistent colors for the same categories across figures
Use appropriate colormaps for data type:
- Sequential: 'viridis', 'plasma' for continuous data
- Diverging: 'RdBu', 'coolwarm' for data with meaningful center
- Qualitative: 'Set1', 'tab10' for categorical data

使用色盲友好的调色板（如'viridis'、'plasma'、'cividis'）
避免使用红绿色组合以保障可访问性
在不同图表中为同一类别使用一致的颜色
根据数据类型选择合适的颜色映射：
- 连续型：'viridis'、'plasma' 适用于连续数据
- 发散型：'RdBu'、'coolwarm' 适用于有意义中心值的数据
- 定性型：'Set1'、'tab10' 适用于分类数据

Figure Layout and Composition

图表布局与构图

Subplots

子图

Use
```
plt.subplots(nrows, ncols)
```
for grid layouts
Use
```
gridspec
```
for complex, non-uniform layouts
Share axes with
```
sharex=True
```
,
```
sharey=True
```
for comparison
Use
```
constrained_layout=True
```
or
```
tight_layout()
```
to prevent overlap

使用
```
plt.subplots(nrows, ncols)
```
创建网格布局
使用
```
gridspec
```
创建复杂的非均匀布局
使用
```
sharex=True
```
、
```
sharey=True
```
共享轴以方便比较
使用
```
constrained_layout=True
```
或
```
tight_layout()
```
防止元素重叠

Figure Size and Resolution

图表尺寸与分辨率

Set figure size explicitly:
```
figsize=(width, height)
```
in inches
Use appropriate DPI for intended output (72 screen, 300+ print)
Standard sizes: (10, 6) for presentations, (8, 6) for papers

显式设置图表尺寸：
```
figsize=(width, height)
```
，单位为英寸
根据预期输出选择合适的DPI（屏幕用72，印刷用300+）
标准尺寸：演示文稿用(10, 6)，论文用(8, 6)

Customization

定制化

Style Sheets

样式表

Use built-in styles:
```
plt.style.use('seaborn-v0_8')
```
,
```
'ggplot'
```
Create custom style files for consistent branding

Combine styles:

plt.style.use(['seaborn-v0_8', 'custom.mplstyle'])

使用内置样式：
```
plt.style.use('seaborn-v0_8')
```
、
```
'ggplot'
```
创建自定义样式文件以保持品牌一致性

组合样式：

plt.style.use(['seaborn-v0_8', 'custom.mplstyle'])

Text and Fonts

文本与字体

Use LaTeX for mathematical notation:
```
r'$\alpha = \frac{1}{2}$'
```
Set font family for consistency
Adjust font sizes for readability at intended display size

使用LaTeX表示数学符号：
```
r'$\alpha = \frac{1}{2}$'
```
设置统一的字体族
根据预期显示尺寸调整字体大小以保障可读性

Saving and Exporting

保存与导出

File Formats

文件格式

Use vector formats (PDF, SVG, EPS) for publications
Use PNG for web and presentations with transparency
Use JPEG only for photographs (lossy compression)

出版用矢量格式（PDF、SVG、EPS）
网页和演示文稿用PNG（支持透明）
仅在处理照片时使用JPEG（有损压缩）

Export Settings

导出设置

Use
```
bbox_inches='tight'
```
to remove excess whitespace
Set
```
facecolor
```
for background color
Specify
```
dpi
```
appropriate for use case
Use
```
transparent=True
```
for overlays

使用
```
bbox_inches='tight'
```
去除多余空白
设置
```
facecolor
```
指定背景色
根据使用场景指定
```
dpi
```
使用
```
transparent=True
```
创建叠加层

Performance Optimization

性能优化

Use
```
rasterized=True
```
for scatter plots with many points
Consider downsampling data for visualization
Close figures with
```
plt.close()
```
after saving
Use
```
plt.close('all')
```
in loops creating many figures

对包含大量点的散点图使用
```
rasterized=True
```
考虑对数据进行下采样后再可视化
保存后使用
```
plt.close()
```
关闭图表
在循环创建大量图表时使用
```
plt.close('all')
```

Key Conventions

关键约定

Import as
```
import matplotlib.pyplot as plt
```
Use object-oriented API for production code
Always label axes and include units
Test visualizations at intended display size
Consider accessibility in color choices

导入方式为
```
import matplotlib.pyplot as plt
```
生产代码使用面向对象API
始终为轴添加标签并包含单位
在预期显示尺寸下测试可视化效果
颜色选择时考虑可访问性