Python - NumPy

Card Puncher Data Processing

About

NumPy is a Python library for working with arrays.

NumPy provides abstractions that make it easy to treat these underlying arrays as vectors and matrices.

Initialization

Installation

Import

# It is convention to import NumPy with the alias np
import numpy as np

Vector Operations

Vector construction

The building block for NumPy is the ndarray, which is a multidimensional array of fixed-size that contains elements of one type (e.g. array of floats).

A numpy array with the values 1, 2, 3

simpleArray = np.array([1, 2, 3])
print simpleArray
[1 2 3]

Combine

Horizontal

np.hstack

>>> a = np.array((1,2,3))
>>> b = np.array((2,3,4))
>>> np.hstack((a,b))
array([1, 2, 3, 2, 3, 4])
>>> a = np.array([[1],[2],[3]])
>>> b = np.array([[2],[3],[4]])
>>> np.hstack((a,b))
array([[1, 2],
       [2, 3],
       [3, 4]])

Vertical

np.vstack

>>> a = np.array([1, 2, 3])
>>> b = np.array([2, 3, 4])
>>> np.vstack((a,b))
array([[1, 2, 3],
       [2, 3, 4]])

Scalar Product

timesFive = 5*simpleArray
[ 5 10 15]

Element-wise multiplication

Element-wise multiplication occurs automatically when you use the * operator to multiply two ndarray objects of the same length.

u = np.arange(0, 5, .5)
v = np.arange(5, 10, .5)

elementWise = u*v
print 'u: {0}'.format(u)
print 'v: {0}'.format(v)
print '\nelementWise\n{0}'.format(elementWise)

u: [ 0.   0.5  1.   1.5  2.   2.5  3.   3.5  4.   4.5]
v: [ 5.   5.5  6.   6.5  7.   7.5  8.   8.5  9.   9.5]

elementWise
[  0.     2.75   6.     9.75  14.    18.75  24.    29.75  36.    42.75]

Dot product

To perform the dot product you can use either np.dot() or np.ndarray.dot().

For example, if you had NumPy arrays x and y, you could compute their dot product four ways:

np.dot(x, y)
np.dot(y, x)
x.dot(y)
y.dot(x)

Example:

u = np.arange(0, 5, .5)
v = np.arange(5, 10, .5)

dotProduct = u.dot(v)
print 'u: {0}'.format(u)
print 'v: {0}'.format(v)

u: [ 0.   0.5  1.   1.5  2.   2.5  3.   3.5  4.   4.5]
v: [ 5.   5.5  6.   6.5  7.   7.5  8.   8.5  9.   9.5]

dotProduct
183.75

Matrix Operations

You can perform matrix math on NumPy matrices using *.

Matrix Construction

You can use np.matrix() to generate a NumPy matrix. Just pass a two-dimensional ndarray or a list of lists to the function. You can perform matrix math on NumPy matrices using *.

A = np.matrix([[1,2,3,4],[5,6,7,8]])
print 'A:\n{0}'.format(A)

A:
[[1 2 3 4]
 [5 6 7 8]]

Transpose

You can transpose a matrix :

# Print A transpose
print '\nA transpose:\n{0}'.format(A.T)

A transpose:
[[1 5]
 [2 6]
 [3 7]
 [4 8]]

Inverting

Inverting a matrix can be done using numpy.linalg.inv().

# Multiply A by A transpose
AAt = A * A.T
print '\nAAt:\n{0}'.format(AAt)

AAt:
[[ 30  70]
 [ 70 174]]

# Invert AAt with np.linalg.inv()
from numpy.linalg import inv
AAtInv = np.linalg.inv(A * A.T)
print '\nAAtInv:\n{0}'.format(AAtInv)

AAtInv:
[[ 0.54375 -0.21875]
 [-0.21875  0.09375]]

# Show inverse times matrix equals identity
# We round due to numerical precision
print '\nAAtInv * AAt:\n{0}'.format((AAtInv * AAt).round(4))

AAtInv * AAt:
[[ 1.  0.]
 [-0.  1.]]





Discover More
Card Puncher Data Processing
Python - Wheel Archive (whl - bdist_wheel)

Wheel is a compiled archive format with the .whl extension. A wheel is a built package that can be installed without needing to go through the “build” process. Installing wheels is substantially faster...
Card Puncher Data Processing
Spark - Dense Vector

A DenseVector is class within the module pyspark.mllib.linalg....



Share this page:
Follow us:
Task Runner