General-purpose, high-level programming language supporting multiple programming paradigms.

Scientific Python Stack

There are lot of libraries for scientific computation and visualization available in Fedora. Well known and widely used is SciPy Stack which consists of

  • Python, a general purpose object-oriented programming language
  • NumPy, a Python library providing fast multidimensional arrays with vector operations
  • SciPy, a Python library providing computational routines, for example numerical integration, various equation solvers and optimization
  • matplotlib, a powerful Python library providing scientific visualization of publication quality
  • IPython, an enhanced interactive Python interpreter
  • pandas, a Python library providing data type for data series manipulation
  • SymPy, a Python library for computer algebra support (i.e. symbolical computation)
  • Jupyter Notebook, a web app that allows you to create and share live code, equations, visualizations and explanatory text


NumPy is a high performance Python library providing fast multidimensional arrays featuring vector operations.

NumPy can be installed in Fedora by typing:

$ sudo dnf install python3-numpy

What are vector operations? The ones which are applied on multiple values at once:

>>> import numpy as np
>>> A = np.array([[1, 2], [3, 4]])
>>> A
array([[1, 2],
       [3, 4]])
>>> A+1
array([[2, 3],
       [4, 5]])
>>> -1*A
array([[-1, -2],
       [-3, -4]])
>>> A**2
array([[ 1,  4],
       [ 9, 16]])
>>> A[0,:]
array([1, 2])
>>> A[0,:]=0
>>> A
array([[0, 0],
       [3, 4]])

NumPy also provides additional mathematical functions like sin, cos, arcsin, exp, log, min, max, sum and others.

>>> np.sin([0, np.pi/6, np.pi/2])
array([ 0. ,  0.5,  1. ])
>>> 2*np.arcsin([0, 1])
array([ 0.        ,  3.14159265])

NumPy includes some handy shortcuts:

>>> np.zeros((3, 3))
array([[ 0.,  0.,  0.],
       [ 0.,  0.,  0.],
       [ 0.,  0.,  0.]])
>>> np.ones((3, 3))
array([[ 1.,  1.,  1.],
       [ 1.,  1.,  1.],
       [ 1.,  1.,  1.]])
>>> np.ones((3, 1))
array([[ 1.],
       [ 1.],
       [ 1.]])
>>> np.zeros_like(np.ones((1, 3)))
array([[ 0.,  0.,  0.]])


SciPy is a Python library providing routines for basic and special mathematical functions, numerical integration, optimization, interpolation, Fourier transform, signal processing, routines for linear algebra, statistics and others. See a full SciPy reference.

To install it, type:

$ sudo dnf install python3-scipy

To compute, for example, the definite integral of 2x from 0 to 1 (which equals 1), type:

>>> import scipy.integrate
>>> scipy.integrate.quad(lambda x: 2*x, 0, 1)
(1.0, 1.1102230246251565e-14)


matplotlib is a graph plotting library producing publication-quality plots.

To install it, type:

$ sudo dnf install python3-matplotlib

Test it by typing:

>>> import matplotlib.pyplot as plt
>>> plt.ion()
>>> plt.plot([1, 2, 3], [10, 20, 30], 'ro--')
>>> plt.title("Hello, matplotlib!")

You should see the plot already.

See lots of other examples at a showcases gallery:

histogram subplot example

3dplot example


IPython is a rich Python interpreter aiming at high-quality user experience for interactive computing and data visualization. Its main features are a tab completion, integration of commands for filesystem access, object introspection and others. IPython provides the kernel for the Jupyter project.

To install and run IPython, type:

$ sudo dnf install python3-ipython
$ ipython3
Python 3.5.1 (default, Mar  4 2016, 15:21:15)
Type "copyright", "credits" or "license" for more information.

IPython 3.2.1 -- An enhanced Interactive Python.
?         -> Introduction and overview of IPython's features.
%quickref -> Quick reference.
help      -> Python's own help system.
object?   -> Details about 'object', use 'object??' for extra details.

In [1]: print("Hello, world!")
Hello, world!


Pandas is a Python library implementing a data type to store data series and relevant routines:

$ sudo dnf install python3-pandas

Let us load famous Iris dataset, which is used a lot as an example in pattern recognition and machine learning literature.

import pandas as pd

iris_data_set_url = (""

# pandas likes CSV files, it can read them from file or provided URL
my_first_pandas_DataFrame = pd.read_csv(iris_data_set_url, header=None)

# See!
print("Pandas DataFrame:")

# Let us find a record with maximal value in the first column

See more on its usage in a pandas tutorial.


SymPy extends SciPy with symbolic computation capabilities, i.e. manipulation with algebraic variables, symbols and precise values.

First, install SymPy by running:

$ sudo dnf install python3-sympy

and then your journey to the realm of CAS (Computer Algebra System) can begin:

>>> import sympy
>>> sympy.sqrt(8)
>>> from sympy import symbols, exp, integrate, oo, diff
>>> x = symbols('x')
>>> integrate(exp(-x), (x, 0, +oo))
>>> integrate(exp(-x), (x, -oo, +oo))
>>> diff(exp(-x))

Jupyter Notebook

Run and share live code via the browser; uses include: data cleaning and transformation, numerical simulation, statistical modeling, machine learning.

First install Jupyter Notebook:

$ sudo dnf install notebook

To start a notebook server run:

$ jupyter notebook

Next select the dropdown New then Python 3 to create a new notebook.

notebook example

NumPy, Pandas, SciPy and IPython are included with the install of Jupyter Notebook.

What next?