python 2d array

Question

How to define a 2-D array without an initialized length, something like this: A = [][]

Answer 1

You can use NumPy library of Python, below I have mentioned few ways to create 2-D arrays using NumPy:
>>> import numpy
>>> numpy.zeros((5, 5))
array([[ 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0.]])
numpy provides a matrix type as well. It's less commonly used, and some people recommend against using it. But it's useful for people coming to numpy from Matlab, and in some other contexts. I thought I'd include it since we're talking about matrices!

>>> numpy.matrix([[1, 2], [3, 4]])
matrix([[1, 2],
[3, 4]])
Here are some other ways to create 2-d arrays and matrices (with output removed for compactness):

numpy.matrix('1 2; 3 4') # use Matlab-style syntax
numpy.arange(25).reshape((5, 5)) # create a 1-d range and reshape
numpy.array(range(25)).reshape((5, 5)) # pass a Python range and reshape
numpy.array([5] * 25).reshape((5, 5)) # pass a Python list and reshape
numpy.empty((5, 5)) # allocate, but don't initialize
numpy.ones((5, 5)) # initialize with ones
numpy.ndarray((5, 5)) # use the low-level constructor

Answer 2

>>> import numpy
>>> numpy.zeros((5, 5))
array([[ 0.,  0.,  0.,  0.,  0.],
       [ 0.,  0.,  0.,  0.,  0.],
       [ 0.,  0.,  0.,  0.,  0.],
       [ 0.,  0.,  0.,  0.,  0.],
       [ 0.,  0.,  0.,  0.,  0.]])

Answer 3

>>> matrix = 5*[5*[0]]
>>> matrix
[[0, 0, 0, 0, 0], [0, 0, 0, 0, 0], [0, 0, 0, 0, 0], [0, 0, 0, 0, 0], [0, 0, 0, 0, 0]]
>>> matrix[4][4] = 2
>>> matrix
[[0, 0, 0, 0, 2], [0, 0, 0, 0, 2], [0, 0, 0, 0, 2], [0, 0, 0, 0, 2], [0, 0, 0, 0, 2]]

Answer 4

You should make a list of lists, and the best way is to use nested comprehensions:

>>> matrix = [[0 for i in range(5)] for j in range(5)]
>>> pprint.pprint(matrix)
[[0, 0, 0, 0, 0],
 [0, 0, 0, 0, 0],
 [0, 0, 0, 0, 0],
 [0, 0, 0, 0, 0],
 [0, 0, 0, 0, 0]]