scipy-best-practices

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

SciPy Best Practices

SciPy 最佳实践

Expert guidelines for SciPy development, focusing on scientific computing, optimization, signal processing, and statistical analysis.

针对SciPy开发的专家指南，聚焦于科学计算、优化、信号处理与统计分析。

Code Style and Structure

代码风格与结构

Write concise, technical Python code with accurate SciPy examples
Prioritize numerical accuracy and computational efficiency
Use functional programming patterns for mathematical operations
Prefer vectorized operations over explicit loops
Use descriptive variable names reflecting scientific context
Follow PEP 8 style guidelines

编写简洁、专业的Python代码，并附带准确的SciPy示例
优先考虑数值准确性与计算效率
对数学运算使用函数式编程模式
优先使用向量化运算而非显式循环
使用反映科学场景的描述性变量名
遵循PEP 8风格指南

scipy.optimize - Optimization

scipy.optimize - 优化

Use
```
scipy.optimize.minimize()
```
for general-purpose optimization
Choose appropriate method based on problem characteristics:
- ```
'BFGS'
```
  for smooth, unconstrained problems
- ```
'L-BFGS-B'
```
  for bounded problems
- ```
'SLSQP'
```
  for constrained optimization
- ```
'Nelder-Mead'
```
  for non-differentiable functions
Provide gradients when available for faster convergence
Use
```
scipy.optimize.curve_fit()
```
for nonlinear least squares fitting
Use
```
scipy.optimize.root()
```
for finding roots of equations

使用
```
scipy.optimize.minimize()
```
进行通用优化
根据问题特性选择合适的方法：
- ```
'BFGS'
```
  用于平滑、无约束问题
- ```
'L-BFGS-B'
```
  用于有界问题
- ```
'SLSQP'
```
  用于约束优化
- ```
'Nelder-Mead'
```
  用于不可微函数
若有梯度信息则提供，以加快收敛速度
使用
```
scipy.optimize.curve_fit()
```
进行非线性最小二乘拟合
使用
```
scipy.optimize.root()
```
求解方程的根

scipy.linalg - Linear Algebra

scipy.linalg - 线性代数

Prefer
```
scipy.linalg
```
over
```
numpy.linalg
```
for additional functionality
Use
```
scipy.linalg.solve()
```
instead of computing matrix inverse
Leverage specialized solvers for structured matrices (banded, triangular)
Use
```
scipy.linalg.lu_factor()
```
and
```
lu_solve()
```
for multiple right-hand sides
Use sparse matrix solvers from
```
scipy.sparse.linalg
```
for large sparse systems

优先使用
```
scipy.linalg
```
而非
```
numpy.linalg
```
，以获得更多功能
使用
```
scipy.linalg.solve()
```
而非计算矩阵逆
针对结构化矩阵（带状、三角矩阵）使用专用求解器
使用
```
scipy.linalg.lu_factor()
```
和
```
lu_solve()
```
处理多右侧项问题
针对大型稀疏系统，使用
```
scipy.sparse.linalg
```
中的稀疏矩阵求解器

scipy.stats - Statistics

scipy.stats - 统计

Use distribution objects for probability calculations
Leverage
```
scipy.stats.describe()
```
for summary statistics

Use hypothesis testing functions:

ttest_ind()

chi2_contingency()

mannwhitneyu()

Generate random samples with
```
.rvs()
```
method on distributions
Use
```
.fit()
```
for parameter estimation from data

使用分布对象进行概率计算
利用
```
scipy.stats.describe()
```
生成汇总统计信息

使用假设检验函数：

ttest_ind()

、

chi2_contingency()

、

mannwhitneyu()

调用分布对象的
```
.rvs()
```
方法生成随机样本
使用
```
.fit()
```
从数据中估计参数

scipy.interpolate - Interpolation

scipy.interpolate - 插值

Use
```
scipy.interpolate.interp1d()
```
for 1D interpolation
Use
```
scipy.interpolate.griddata()
```
for scattered data interpolation
Choose appropriate interpolation method: 'linear', 'cubic', 'nearest'
Use spline functions for smooth interpolation:
```
UnivariateSpline
```
,
```
BSpline
```
Consider
```
RegularGridInterpolator
```
for regular grid data

使用
```
scipy.interpolate.interp1d()
```
进行一维插值
使用
```
scipy.interpolate.griddata()
```
进行散点数据插值
选择合适的插值方法：'linear'、'cubic'、'nearest'
使用样条函数实现平滑插值：
```
UnivariateSpline
```
、
```
BSpline
```
针对规则网格数据，考虑使用
```
RegularGridInterpolator
```

scipy.integrate - Integration

scipy.integrate - 积分

Use
```
scipy.integrate.quad()
```
for single integrals
Use
```
scipy.integrate.dblquad()
```
,
```
tplquad()
```
for multiple integrals
Use
```
scipy.integrate.solve_ivp()
```
for ordinary differential equations
Choose appropriate ODE method: 'RK45', 'BDF', 'LSODA'
Provide Jacobian for stiff systems to improve performance

使用
```
scipy.integrate.quad()
```
进行单积分
使用
```
scipy.integrate.dblquad()
```
、
```
tplquad()
```
进行多重积分
使用
```
scipy.integrate.solve_ivp()
```
求解常微分方程
选择合适的ODE方法：'RK45'、'BDF'、'LSODA'
针对刚性系统提供雅可比矩阵以提升性能

scipy.signal - Signal Processing

scipy.signal - 信号处理

Use
```
scipy.signal.butter()
```
,
```
cheby1()
```
,
```
ellip()
```
for filter design
Apply filters with
```
scipy.signal.filtfilt()
```
for zero-phase filtering
Use
```
scipy.signal.welch()
```
for power spectral density estimation
Use
```
scipy.signal.find_peaks()
```
for peak detection
Leverage
```
scipy.signal.convolve()
```
and
```
correlate()
```
for convolution

使用
```
scipy.signal.butter()
```
、
```
cheby1()
```
、
```
ellip()
```
进行滤波器设计
使用
```
scipy.signal.filtfilt()
```
应用滤波器以实现零相位滤波
使用
```
scipy.signal.welch()
```
估计功率谱密度
使用
```
scipy.signal.find_peaks()
```
进行峰值检测
利用
```
scipy.signal.convolve()
```
和
```
correlate()
```
进行卷积运算

scipy.sparse - Sparse Matrices

scipy.sparse - 稀疏矩阵

Use appropriate sparse format for your use case:
- ```
csr_matrix
```
  for efficient row slicing and matrix-vector products
- ```
csc_matrix
```
  for efficient column slicing
- ```
coo_matrix
```
  for constructing sparse matrices
- ```
lil_matrix
```
  for incremental construction
Convert to optimal format before operations
Use
```
scipy.sparse.linalg
```
solvers for sparse linear systems

根据使用场景选择合适的稀疏格式：
- ```
csr_matrix
```
  用于高效行切片和矩阵-向量乘积
- ```
csc_matrix
```
  用于高效列切片
- ```
coo_matrix
```
  用于构建稀疏矩阵
- ```
lil_matrix
```
  用于增量构建矩阵
在执行运算前转换为最优格式
使用
```
scipy.sparse.linalg
```
求解器处理稀疏线性系统

Performance Optimization

性能优化

Use appropriate data types (
```
float64
```
for precision,
```
float32
```
for memory)
Leverage BLAS/LAPACK through SciPy for optimized linear algebra
Pre-allocate arrays when possible
Use in-place operations when available

使用合适的数据类型（
```
float64
```
保证精度，
```
float32
```
节省内存）
通过SciPy利用BLAS/LAPACK实现优化的线性代数运算
尽可能预先分配数组
尽可能使用原地操作

Error Handling and Validation

错误处理与验证

Check convergence status of optimization routines
Validate numerical results for reasonableness
Handle ill-conditioned problems gracefully
Use appropriate tolerances for convergence criteria

检查优化例程的收敛状态
验证数值结果的合理性
优雅处理病态问题
为收敛准则设置合适的容差

Testing Scientific Code

科学代码测试

Test against known analytical solutions
Use
```
np.testing.assert_allclose()
```
for numerical comparisons
Test edge cases and boundary conditions
Verify conservation laws and invariants

对照已知的解析解进行测试
使用
```
np.testing.assert_allclose()
```
进行数值比较
测试边缘情况与边界条件
验证守恒律与不变量

Key Conventions

关键约定

Import specific submodules:

from scipy import optimize, stats, linalg

Use
```
snake_case
```
for variables and functions
Document algorithm choices and parameters
Include convergence diagnostics in output

导入特定子模块：

from scipy import optimize, stats, linalg

变量和函数使用
```
snake_case
```
命名法
记录算法选择与参数
在输出中包含收敛诊断信息