[FEATURE REQUEST] numpy.insert()

[water5]

https://numpy.org/doc/1.21/reference/generated/numpy.insert.html

b = np.array([[1, 2, 3], [4, 5, 6]])

array([[1, 2, 3], [4, 5, 6]])

np.insert(b, 1, [7, 8, 9])

array([1, 7, 8, 9, 2, 3, 4, 5, 6])

np.insert(b, 1, [7, 8, 9], axis = 0)

array([[1, 2, 3], [7, 8, 9], [4, 5, 6]])

[v923z]

@water5 Wouldn't this basically duplicate the functionality of concatenate? https://micropython-ulab.readthedocs.io/en/latest/ulab-ndarray.html?highlight=concatenate#concatenate

[water5]

But that can't specify where to insert, I consider numpy.append before post this request.

[v923z]

Obviously, you first have to split your original array into two with slicing (that is, where you specify where you want to insert the values), and then concatenate the first slice, the value, and the second slice.

a = array([1, 2, 3, 4, 5, 6])
value = array([10, 11])
concatenate(a[:3], value, a[3:])

[v923z]

Append is similar

a = array([1, 2, 3, 4, 5, 6])
b = array([10, 20, 30])
c = zeros(len(a) + len(b))
c[:len(a)] = a
c[len(a):] = b

[water5]

On 2D array,

a = np.arange(25).reshape(5, 5)
a

array([[ 0, 1, 2, 3, 4], [ 5, 6, 7, 8, 9], [10, 11, 12, 13, 14], [15, 16, 17, 18, 19], [20, 21, 22, 23, 24]])

b = np.arange(25, 31)
b

array([25, 26, 27, 28, 29])

np.insert(a, 1, b, axis = 0)

array([[ 0, 1, 2, 3, 4], [25, 26, 27, 28, 29], [ 5, 6, 7, 8, 9], [10, 11, 12, 13, 14], [15, 16, 17, 18, 19], [20, 21, 22, 23, 24]])

np.insert(a, 1, b, axis = 1)

array([[ 0, 25, 1, 2, 3, 4], [ 5, 26, 6, 7, 8, 9], [10, 27, 11, 12, 13, 14], [15, 28, 16, 17, 18, 19], [20, 29, 21, 22, 23, 24]])

np.concatenate(a[ : 1], b, a[1 : ])

Traceback (most recent call last): File "<stdin>", line 1, in <module> File "<__array_function__ internals>", line 180, in concatenate TypeError: only integer scalar arrays can be converted to a scalar index

[v923z]

a = array(range(25)).reshape((5,5))
b = np.arange(25, 30).reshape((1,5))
np.concatenate((a[:1], b, a[1:]))

[water5]

Right, that seems could implement most numpy.insert function. I close this issue directly, or make a snippet?

[v923z]

Right, that seems could implement most numpy.insert function. I close this issue directly, or make a snippet?

You could definitely write a small snippet, if you feel like it. In that case, add the snippet label to the issue, and keep it open till you are done, otherwise, close it.