Expand - 8 vs 13#
Next section compares an older to a newer version of the same operator after both definition are converted into markdown text. Green means an addition to the newer version, red means a deletion. Anything else is unchanged.
- Expand8 → Expand13 +0 -1
Expand8 → Expand13
RENAMED
|
@@ -1 +1 @@
|
|
|
1
1
|
Broadcast the input tensor following the given shape and the broadcast rule.
|
|
2
2
|
The broadcast rule is similar to numpy.array(input) * numpy.ones(shape):
|
|
3
3
|
Dimensions are right alignment;
|
|
4
4
|
Two corresponding dimensions must have the same value, or one of them is equal to 1.
|
|
5
5
|
Also, this operator is similar to numpy.broadcast_to(input, shape),
|
|
6
6
|
but the major difference is numpy.broadcast_to() does not allow shape to be smaller than input.size().
|
|
7
7
|
It is possible that the output.shape is not equal to shape, when some dimensions in shape is equal to 1,
|
|
8
8
|
or the shape.ndim < input.shape.ndim.
|
|
9
9
|
**Inputs**
|
|
10
10
|
* **input** (heterogeneous) - **T**:
|
|
11
11
|
Input tensor
|
|
12
12
|
* **shape** (heterogeneous) - **tensor(int64)**:
|
|
13
13
|
A 1-D tensor indicates the shape you want to expand to, following
|
|
14
14
|
the broadcast rule
|
|
15
15
|
**Outputs**
|
|
16
16
|
* **output** (heterogeneous) - **T**:
|
|
17
17
|
Output tensor
|
|
18
18
|
**Type Constraints**
|
|
19
19
|
* **T** in (
|
|
20
|
-
tensor(bfloat16),
|
|
21
20
|
tensor(bool),
|
|
22
21
|
tensor(complex128),
|
|
23
22
|
tensor(complex64),
|
|
24
23
|
tensor(double),
|
|
25
24
|
tensor(float),
|
|
26
25
|
tensor(float16),
|
|
27
26
|
tensor(int16),
|
|
28
27
|
tensor(int32),
|
|
29
28
|
tensor(int64),
|
|
30
29
|
tensor(int8),
|
|
31
30
|
tensor(string),
|
|
32
31
|
tensor(uint16),
|
|
33
32
|
tensor(uint32),
|
|
34
33
|
tensor(uint64),
|
|
35
34
|
tensor(uint8)
|
|
36
35
|
):
|
|
37
36
|
Constrain input and output types to all tensors.
|