[FEATURE REQUEST] numpy.unique

[water5]

https://numpy.org/doc/stable/reference/generated/numpy.unique.html

[v923z]

While it is not the most performant way, I think you can already get this functionality by

a = np.array(...)
set(a.tolist())

[water5]

But set not support ndarray to list result.

a = np.arange(9).reshape((3, 3))
a

array([[0, 1, 2], [3, 4, 5], [6, 7, 8]], dtype=int16)

a = a.tolist()
a

[[0, 1, 2], [3, 4, 5], [6, 7, 8]]

set(a)

Traceback (most recent call last): File "<stdin>", line 1, in <module> TypeError: unsupported type for __hash__: 'list'

Only support 1D:

a = np.arange(9)
a

array([0, 1, 2, 3, 4, 5, 6, 7, 8], dtype=int16)

set(a)

{0, 1, 2, 3, 4, 5, 6, 7, 8}

[water5]

Although we can do set operation after numpy.reshape:

a = a.reshape((9, ))
set(a)
a = a.reshape((3, 3))

But the return_index parameter result will pointless.

[v923z]

I am trying to understand in what situation this function would be required.