CODEWITHSUNDEEP
pythonnumpy
London guy
London guy

Reputation: 28022

What does this parameter in numpy.zeros method mean?

I have this piece of code and I am finding it difficult to understand what is advantage of defining the numpy.zeros method the way it is as shown below.

Z = np.zeros((10,10), [('x',float),('y',float)])
Z['x'], Z['y'] = np.meshgrid(np.linspace(0,1,10),
                         np.linspace(0,1,10))

print(Z)

What is the significance of mentioning of x and y?

Upvotes: 3

Views: 2641

Answers (3)

hpaulj
hpaulj

Reputation: 231375

The zeros creates a (10,10) array, where each element has dtype defined by np.dtype([('x',float),('y',float)]). That is, each element consists of 2 floats, one called 'x', the other 'y'.

Z = np.zeros((10,10), [('x',float),('y',float)])

In a sense this makes a (10,10,2) array, except that there is a 'wall' between the 2 dimension, and the others. For example you can't swap it one other other dimensions. But it is possible to 'view' it as a (10,10,2) array:

Z.view('float').reshape(10,10,2)

The 2 fields of Z are indexed with Z['x'] and Z['y'], the resulting views are each (10,10) arrays.

The 2nd line sets the values of these 2 fields

Z['x'], Z['y'] = np.meshgrid ...

Normally meshgrid returns 2 arrays, X, Y = np.meshgrid.... So this is just a normal Python assignment.

I haven't seen this pairing of a structured array and meshgrid before, but it makes sense. Whether it is all that useful in another matter.

I was going to add an example of what Z looks like, but @AndreL has done that for us. Note that the elements Z are displayed as tuples, implying they are different from 2 element columns of a 3d array.

Upvotes: 3

Shapi
Shapi

Reputation: 5593

The secret of the output is at numpy.linspace(0,1,10), outputs an numpy.array, with: 

[ 0.          0.11111111  0.22222222  0.33333333  0.44444444  0.55555556
  0.66666667  0.77777778  0.88888889  1.        ]

For 'x' shape, as for 'y', where '0' is where it starts, '1' is where is stops, with 10 samples.

The numpy.zeros() are defining a matrix shape (M, N) for ‘ij’ indexing, where M = N = 10

The numpy.meshgrid() indexes into the matrix the values of linspace results, like ai, aj

e.g.

Z = np.zeros((10,10), [('x',int),('y',int)])
Z['x'], Z['y'] = np.meshgrid( np.linspace(0,10,10), np.linspace(0,10,10))
print Z

Outputs:

[[(0, 0) (1, 0) (2, 0) (3, 0) (4, 0) (5, 0) (6, 0) (7, 0) (8, 0) (10, 0)]
 [(0, 1) (1, 1) (2, 1) (3, 1) (4, 1) (5, 1) (6, 1) (7, 1) (8, 1) (10, 1)]
 [(0, 2) (1, 2) (2, 2) (3, 2) (4, 2) (5, 2) (6, 2) (7, 2) (8, 2) (10, 2)]
 [(0, 3) (1, 3) (2, 3) (3, 3) (4, 3) (5, 3) (6, 3) (7, 3) (8, 3) (10, 3)]
 [(0, 4) (1, 4) (2, 4) (3, 4) (4, 4) (5, 4) (6, 4) (7, 4) (8, 4) (10, 4)]
 [(0, 5) (1, 5) (2, 5) (3, 5) (4, 5) (5, 5) (6, 5) (7, 5) (8, 5) (10, 5)]
 [(0, 6) (1, 6) (2, 6) (3, 6) (4, 6) (5, 6) (6, 6) (7, 6) (8, 6) (10, 6)]
 [(0, 7) (1, 7) (2, 7) (3, 7) (4, 7) (5, 7) (6, 7) (7, 7) (8, 7) (10, 7)]
 [(0, 8) (1, 8) (2, 8) (3, 8) (4, 8) (5, 8) (6, 8) (7, 8) (8, 8) (10, 8)]
 [(0, 10) (1, 10) (2, 10) (3, 10) (4, 10) (5, 10) (6, 10) (7, 10) (8, 10)
  (10, 10)]]

Outputting a matrix ij scalars.

Check the next url's:

  1. numpy.linspace()
  2. numpy.zeros()
  3. numpy.meshgrid()

Upvotes: 1

DilithiumMatrix
DilithiumMatrix

Reputation: 18627

This is actually defining two separate ndarrays, one named 'x' and the other 'y'. While, in this case, it is unnecessary to specify the dtypes, it is a way of creating this type of double ndarray.

While this usage is not explicitly included in the numpy.zeros documentation, they do show an example using it.

Edit:

@WarrenWeckesser links some documentation for these structured arrays

Upvotes: 3

Related Questions