I have a 3D nested list constructed like this:
[[[1,1],[2,1],[2,2],[1,2]],[[-1,-1],[-2,-1],[-2,-2],[-1,-2]]]
and I want to concat to that [[0,0]] to get this result:
[[[1,1],[2,1],[2,2],[1,2]],[[-1,-1],[-2,-1],[-2,-2],[-1,-2]],[[0,0]]]
I tried to use numpy.append and column_stack methods, but it complains about dimension mismatch. How to do that right?
The problem is that once you have a numpy array with shape (2, 4, 2), you cannot really append another array with shape (1, 1, 2) unless you treat lists as objects (i.e. you are appending an element to an array of size 2 so that in the end you have an array of size 3).
If you have a numpy array of list objects you won't be able to benefit from most of the functionality provided by numpy so perhaps explain in a bit more detail what exactly you are trying to achieve here.
>>> a = np.array([[[1,1],[2,1],[2,2],[1,2]],[[-1,-1],[-2,-1],[-2,-2],[-1,-2]]])
>>> b = np.array([[[0, 0]]])
>>> a.shape, b.shape
((2, 4, 2), (1, 1, 2))
>>> c = np.array(a.tolist() + b.tolist())
>>> c
array([list([[1, 1], [2, 1], [2, 2], [1, 2]]),
list([[-1, -1], [-2, -1], [-2, -2], [-1, -2]]), list([[0, 0]])],
dtype=object)
>>> c.shape
(3,)
>>> c.tolist()
[[[1, 1], [2, 1], [2, 2], [1, 2]],
[[-1, -1], [-2, -1], [-2, -2], [-1, -2]],
[[0, 0]]]
The last value is the list you seem to be looking for. You don't have to go through this process with numpy to get it and can simply append as others have suggested.
[[1,1],[2,1],[2,2],[1,2]] and [[-1,-1],[-2,-1],[-2,-2],[-1,-2]] are coordinates of rectangles vertices and [[0,0]] is a point's coordinates, which I want to add to array which has rectangles (for later processing)
guess what, I can just use default append function; I didn't know that function is not returning the new array
numpy functions. If you start with a list, and want a list result, using numpy is often slower, and possibly wrong.
numpy.arrayor a built-inlist?a.append([[0,0]])should work.numpybut possible inlistwithappendmethod.