使用numpy的Python数组

Question

I am confused about doing vectorization using numpy .

In particular, I have a matrix of this form: of type <type 'list'>

[[0.0, 0.0, 0.0, 0.0], [0.02, 0.04, 0.0325, 0.04], [1, 2, 3, 4]]

How do I make it look like the following using numpy?

[[  0.0   0.0   0.0     0.0  ]
 [  0.02  0.04  0.0325  0.04 ]
 [  1     2     3       4    ]]

Yes, I know I can do it using:

np.array([[0.0, 0.0, 0.0, 0.0], [0.02, 0.04, 0.0325, 0.04], [1, 2, 3, 4]])

But I have a very long matrix, and I can't just type out each rows like that. How can I handle the case when I have a very long matrix?

Answer 1

This is not a matrix of type list, it is a list that contains lists. You may think of it as matrix, but to Python it is just a list

alist = [[0.0, 0.0, 0.0, 0.0], [0.02, 0.04, 0.0325, 0.04], [1, 2, 3, 4]]

arr = np.array(alist)

works just the same as

arr = np.array([[0.0, 0.0, 0.0, 0.0], [0.02, 0.04, 0.0325, 0.04], [1, 2, 3, 4]])

This creates 2d array, with shape (3,4) and dtype float

In [212]: arr = np.array([[0.0, 0.0, 0.0, 0.0], [0.02, 0.04, 0.0325, 0.04], [1, 2, 3, 4]])

In [213]: arr
Out[213]: 
array([[ 0.    ,  0.    ,  0.    ,  0.    ],
       [ 0.02  ,  0.04  ,  0.0325,  0.04  ],
       [ 1.    ,  2.    ,  3.    ,  4.    ]])
In [214]: print(arr)
[[ 0.      0.      0.      0.    ]
 [ 0.02    0.04    0.0325  0.04  ]
 [ 1.      2.      3.      4.    ]]

Answer 2

Assuming you start with a large array, why not split it into arrays of the right size ( n ):

splitted = [l[i:i + n] for i in range(0, len(array), n)]

and make the matrix from that:

np.array(splitted)

Answer 3

If you're saying you have a list of lists stored in Python object A , all you need to do is call np.array(A) which will return a numpy array using the elements of A . Otherwise, you need to specify what form your data is in right now to clarify how you want to load your data.