convert multidimensional numpy array to list of string

Question

I have a numpy array:

x = np.array([[1.5,1.3],[2.5,1.5]])

What is the most pythonic way to convert this into a list of strings like this:

y = ['1.5 1.3','2.5 1.5']

What I do now is not very effective:

y = []
for i in range(x.shape[0]):
    txt = ''
    for j in range(x.shape[1]):
       txt+=' '%.5e'%x[i,j]
    y.append(txt)

Have tried with:

x = x.astype('|S10').tolist()
y = [' '.join(i) for i in x]

but if I have a lot of decimals in x, it truncates it after the 10th decimal.

Answer 1

You cam simply use list comprehension:

>>> x = np.array([[1.5,1.3],[2.5,1.5]])
>>> y = [' '.join('{:.5e}'.format(col) for col in line) for line in x]
>>> y 
['1.50000e+00 1.30000e+00', '2.50000e+00 1.50000e+00']

You can change '{:.5}' to whatever format you want.

Answer 2

Here's one approach that starts off by converting the input numeric array to a string array with .astype('S4') conversion and then does one level of loop comprehension to join all strings in each row with " ".join(item), where item is each row from the string array, like so -

[" ".join(item) for item in x.astype('S4')]

If you are looking for precision control, you can change the dtype to S6, S12 or S14.

Sample run -

In [9]: x
Out[9]: 
array([[ 0.62047293,  0.02957529,  0.88920602,  0.57581068],
       [ 0.40903378,  0.80748886,  0.83018903,  0.22445871],
       [ 0.87296866,  0.94234112,  0.70001789,  0.99333763],
       [ 0.96689113,  0.35128491,  0.35775966,  0.26734985]])

In [10]: [" ".join(item) for item in x.astype('S4')]
Out[10]: 
['0.62 0.02 0.88 0.57',
 '0.40 0.80 0.83 0.22',
 '0.87 0.94 0.70 0.99',
 '0.96 0.35 0.35 0.26']

In [11]: [" ".join(item) for item in x.astype('S14')]
Out[11]: 
['0.620472934011 0.029575285327 0.889206021736 0.575810682998',
 '0.409033783485 0.807488858152 0.830189034897 0.224458707937',
 '0.872968659668 0.942341118836 0.700017893576 0.993337626766',
 '0.966891127767 0.351284905075 0.357759658063 0.267349854182']