14

I am using a python wrapper to call functions of a c++ dll library. A ctype is returned by the dll library, which I convert to numpy array

score = np.ctypeslib.as_array(score,1)

however, the array has no shape?

score
>>> array(-0.019486344729027664)

score.shape
>>> ()

score[0]
>>> IndexError: too many indices for array

How can I extract a double from the score array?

Thank you.

jpp
  • 159,742
  • 34
  • 281
  • 339
Alejandro Simkievich
  • 3,512
  • 4
  • 33
  • 49
  • 1
    That is a shape; it's shape `()`, aka 0-dimensional. – user2357112 Jan 21 '16 at 18:15
  • thanks a lot. is there any way to extract the double inside the array? I guess that is the question in the end – Alejandro Simkievich Jan 21 '16 at 18:16
  • You can use `float(score)`. But how are you ending up with a 0-d array, i.e. what's the initial type and value of `score`? – Eryk Sun Jan 21 '16 at 18:18
  • Why are you calling `np.ctypeslib.as_array` on this thing? `1` isn't a valid shape, and if there's only one value, why do you want to use `np.ctypeslib.as_array` to retrieve it? Why not go through the normal ctypes interface? – user2357112 Jan 21 '16 at 18:19
  • The `shape` parameter is only used for a pointer, so we know `score` isn't initially a pointer, else passing `shape=1` would be an error. If you pass `as_array` a scalar such as `c_double(-0.19)`, it stores an `__array_interface__` property on the `c_double` type with `shape=()`. However, in NumPy 1.8.2 this actually creates an array with `shape=(1,)`. Maybe in older versions it creates a scalar 'array'. – Eryk Sun Jan 21 '16 at 18:42

1 Answers1

20

You can access the data inside a 0-dimensional array via indexing [()].

For example, score[()] will retrieve the underlying data in your array.

The idiom is in fact consistent:

# x, y, z are 0-dim, 1-dim, 2-dim respectively
x = np.array(1)
y = np.array([1, 2, 3])
z = np.array([[1, 2, 3], [4, 5, 6]])

# use 0-dim, 1-dim, 2-dim tuple indexers respectively
res_x = x[()]      # 1
res_y = y[(1,)]    # 2
res_z = z[(1, 2)]  # 6

Tuples seem unnatural because you don't need to use them explicitly for the 1d and 2d cases, i.e. y[1] and z[1, 2] suffice. That option isn't available for the 0-dim case, so use the zero-length tuple.

jpp
  • 159,742
  • 34
  • 281
  • 339