-
-
Notifications
You must be signed in to change notification settings - Fork 804
/
Copy pathrepl.txt
71 lines (59 loc) · 1.96 KB
/
repl.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
{{alias}}( arrays )
Broadcasts ndarrays to a common shape.
If a provided ndarray has a shape matching the common shape, the function
returns the provided ndarray.
If a provided ndarray has a different (broadcast compatible) shape than the
common shape, the function returns a new (base) ndarray view of the
provided ndarray's data. The view is typically *not* contiguous. As more
than one element of a returned view may refer to the same memory location,
writing to a view may affect multiple elements. If you need to write to a
returned array, copy the array before performing operations which may
mutate elements.
A returned array view is a "base" ndarray, and, thus, a returned array view
does not perform bounds checking or afford any of the guarantees of the
non- base ndarray constructor. The primary intent of this function is to
broadcast ndarray-like objects within internal implementations and to do so
with minimal overhead.
The function throws an error if a provided broadcast-incompatible ndarrays.
Parameters
----------
arrays: ArrayLikeObject<ndarray>
List of ndarrays.
Returns
-------
out: Array<ndarray>
Broadcasted arrays.
Examples
--------
> var x = {{alias:@stdlib/ndarray/array}}( [ [ 1, 2 ], [ 3, 4 ] ] )
<ndarray>
> var sh = x.shape
[ 2, 2 ]
> var y = {{alias:@stdlib/ndarray/zeros}}( [ 3, 2, 2 ] )
<ndarray>
> var out = {{alias}}( [ x, y ] )
[ <ndarray>, <ndarray> ]
// Retrieve the broadcasted "x" array:
> var bx = out[ 0 ]
<ndarray>
> sh = bx.shape
[ 3, 2, 2 ]
// Retrieve broadcasted elements...
> var v = bx.get( 0, 0, 0 )
1
> v = bx.get( 0, 0, 1 )
2
> v = bx.get( 0, 1, 0 )
3
> v = bx.get( 0, 1, 1 )
4
> v = bx.get( 1, 0, 0 )
1
> v = bx.get( 1, 1, 0 )
3
> v = bx.get( 2, 0, 0 )
1
> v = bx.get( 2, 1, 1 )
4
See Also
--------