Skip to content

SciPy

Data & AnalyticsData SciencePython
Official docs ↗GitHub ↗

What it is

SciPy is an open-source Python library used for scientific and technical computing. It builds on NumPy arrays and provides a wide range of algorithms for optimization, integration, interpolation, eigenvalue problems, algebraic equations, differential equations, and signal processing.

SciPy provides modules for optimization, linear algebra, signal and image processing, statistics, integration, and more. It integrates seamlessly with NumPy arrays and allows advanced scientific computations with minimal code.

Installation

pip install scipy

Getting started

The smallest useful thing you can do with it, and what each part means.

Minimizing a function
from scipy import optimize
f = lambda x: (x - 3)**2
result = optimize.minimize(lambda v: f(v[0]), x0=[0])
print(result)
Demonstrates minimizing a simple quadratic function using SciPy's `optimize.minimize` function.
Computing definite integrals
from scipy import integrate
result, error = integrate.quad(lambda x: x**2, 0, 1)
print(result)
Uses `integrate.quad` to compute the definite integral of x² from 0 to 1.

Advanced usage

Where the library earns its place over a simpler alternative.

Solving linear systems
from scipy import linalg
import numpy as np
A = np.array([[3,2],[1,2]])
b = np.array([5,5])
x = linalg.solve(A, b)
print(x)
Uses SciPy’s linear algebra module to solve a system of linear equations.
Signal processing example
from scipy import signal
import numpy as np
b, a = signal.butter(3, 0.05)
z = signal.lfilter(b, a, np.random.randn(1000))
print(z[:10])
Designs a 3rd-order Butterworth filter and applies it to a random signal using `signal.lfilter`.
Interpolation example
from scipy import interpolate
import numpy as np
x = np.arange(5)
y = np.sin(x)
f = interpolate.interp1d(x, y)
print(f(2.5))
Performs linear interpolation between data points using SciPy’s `interp1d` function.

Errors and fixes

The failures you are most likely to hit, and what actually resolves them.

ValueError
Ensure inputs have correct shapes and types expected by the function.
LinAlgError
Check that matrices are square and non-singular when performing linear algebra operations.

Best practices

  • Use NumPy arrays for input data to ensure compatibility with SciPy functions.
  • Check function domains and constraints when using optimization routines.
  • Leverage built-in documentation for each submodule to understand parameter options.
  • Use `scipy.constants` for physical constants to improve code readability and accuracy.
  • Profile computations for large datasets using vectorized operations rather than loops.

Background

Why it exists, and what it was reacting to.

SciPy was initially created in 2001 by Travis Oliphant, Eric Jones, and Pearu Peterson. It was designed to extend NumPy’s capabilities by providing high-level functions for scientific computations. SciPy has since become a core library in the Python scientific ecosystem and is widely used in engineering, physics, machine learning, and data analysis.

Related libraries

Others in Data & Analytics for Python.

← All Python librariesPython guide