view raw
johnhenry johnhenry - 1 year ago 41
Python Question

Delete columns from matrix according to its index

I have a two-dimensional matrix

. For example it could be something like

matrixK = [[0,1,2,3,4],[5,6,7,8,9],[10,11,12,13,14]]

I need to delete certain columns of the matrix, and these columns need to be chosen according to a pattern that depend on their index as columns.
In other words,

for i in range(number_of_columns)
if (i satisfy a certain condtion):
column[i] needs to be deleted.

and as a final result I need to get the same initial
, deprived of the columns that satisfy the condition. What is the best pythonic way to do that?

To make an example, referring to the
defined above as an example, let's consider

for i in range(5):
if (i%2==0):
column[i] needs to be deleted

which should delete columns 2 and 4.


This can be done two different ways depending on whether you need to delete the column in-place in the original matrix or can just replace it with a new one. An example of each way is shown below.
Note that neither requires every row of the matrix to be the same length.

# in-place removal of column in matrix

matrixK = [[0, 1, 2, 3, 4], [5, 6, 7, 8, 9], [10, 11, 12, 13, 14]]

for row in matrixK:
    for i in reversed(range(len(row))):
        if i % 2 == 0:
           del row[i:i+1]

print('matrixK after: {}'.format(matrixK))


# removal of column in matrix by recreating it

matrixK = [[0, 1, 2, 3, 4], [5, 6, 7, 8, 9], [10, 11, 12, 13, 14]]

matrixK = [[row[i] for i in range(len(row)) if i % 2] for row in matrixK]

print('matrixK after: {}'.format(matrixK))

Both output:

matrixK after: [[1, 3], [6, 8], [11, 13]]