SciPy Sparse Data

 What is Sparse Data

Sparse data is data that has mostly unused elements (elements that don't carry any information ).

It can be an array like this one:

[1, 0, 2, 0, 0, 3, 0, 0, 0, 0, 0, 0]




Sparse Data: is a data set where most of the item values are zero.

Dense Array: is the opposite of a sparse array: most of the values are not zero.

In scientific computing, when we are dealing with partial derivatives in linear algebra we will come across sparse data.

---

How to Work With Sparse Data

SciPy has a module, scipy.sparse that provides functions to deal with sparse data.

There are primarily two types of sparse matrices that we use:

CSC - Compressed Sparse Column. For efficient arithmetic, fast column slicing.

CSR - Compressed Sparse Row. For fast row slicing, faster matrix vector products

We will use the CSR matrix in this tutorial.

---

CSR Matrix

We can create CSR matrix by passing an arrray into function scipy.sparse.csr_matrix().

Example

Create a CSR matrix from an array:

import numpy as np
from scipy.sparse import csr_matrix

arr = np.array([0, 0, 0, 0, 0, 1, 1, 0, 2])

print(csr_matrix(arr))

The example above returns:

  (0, 5)	1
  (0, 6)	1
  (0, 8)	2

From the result we can see that there are 3 items with value.

The 1. item is in row 0 position 5 and has the value 1.

The 2. item is in row 0 position 6 and has the value 1.

The 3. item is in row 0 position 8 and has the value 2.

Sparse Matrix Methods

Viewing stored data (not the zero items) with the data property:

Example

import numpy as np
from scipy.sparse import csr_matrix

arr = np.array([[0, 0, 0], [0, 0, 1], [1, 0, 2]])

print(csr_matrix(arr).data)

 [1 1 2]

Counting nonzeros with the count_nonzero() method:

Example

import numpy as np
from scipy.sparse import csr_matrix

arr = np.array([[0, 0, 0], [0, 0, 1], [1, 0, 2]])

print(csr_matrix(arr).count_nonzero())

3

Removing zero-entries from the matrix with the eliminate_zeros() method:

Example

import numpy as np
from scipy.sparse import csr_matrix

arr = np.array([[0, 0, 0], [0, 0, 1], [1, 0, 2]])

mat = csr_matrix(arr)
mat.eliminate_zeros()

print(mat)

 (1, 2)	        1
 (2, 0) 	1
 (2, 2)	        2

Eliminating duplicate entries with the sum_duplicates() method:

Example

Eliminating duplicates by adding them:

import numpy as np
from scipy.sparse import csr_matrix

arr = np.array([[0, 0, 0], [0, 0, 1], [1, 0, 2]])

mat = csr_matrix(arr)
mat.sum_duplicates()

print(mat)

  (1, 2)	1
  (2, 0)	1
  (2, 2)	2

Converting from csr to csc with the tocsc() method:

Example

import numpy as np
from scipy.sparse import csr_matrix

arr = np.array([[0, 0, 0], [0, 0, 1], [1, 0, 2]])

newarr = csr_matrix(arr).tocsc()

print(newarr)

   (2, 0) 1

  (1, 2) 1

  (2, 2) 2