Pow - 12 vs 15#
Next section compares an older to a newer version of the same operator after both definition are converted into markdown text.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.
- Pow12 → Pow15 +1 -3
Pow12 → Pow15
RENAMED
|
@@ -1 +1 @@
|
|
|
1
1
|
Pow takes input data (Tensor<T>) and exponent Tensor, and
|
|
2
2
|
produces one output data (Tensor<T>) where the function f(x) = x^exponent,
|
|
3
3
|
is applied to the data tensor elementwise.
|
|
4
4
|
This operator supports **multidirectional (i.e., Numpy-style) broadcasting**; for more details please check Broadcasting in ONNX <https://github.com/onnx/onnx/blob/master/docs/Broadcasting.md>_.
|
|
5
5
|
**Inputs**
|
|
6
6
|
* **X** (heterogeneous) - **T**:
|
|
7
7
|
First operand, base of the exponent.
|
|
8
8
|
* **Y** (heterogeneous) - **T1**:
|
|
9
9
|
Second operand, power of the exponent.
|
|
10
10
|
**Outputs**
|
|
11
11
|
* **Z** (heterogeneous) - **T**:
|
|
12
|
-
Output tensor
|
|
12
|
+
Output tensor.
|
|
13
13
|
**Type Constraints**
|
|
14
14
|
* **T** in (
|
|
15
|
-
tensor(bfloat16),
|
|
16
15
|
tensor(double),
|
|
17
16
|
tensor(float),
|
|
18
17
|
tensor(float16),
|
|
19
18
|
tensor(int32),
|
|
20
19
|
tensor(int64)
|
|
21
20
|
):
|
|
22
21
|
Constrain input X and output types to float/int tensors.
|
|
23
22
|
* **T1** in (
|
|
24
|
-
tensor(bfloat16),
|
|
25
23
|
tensor(double),
|
|
26
24
|
tensor(float),
|
|
27
25
|
tensor(float16),
|
|
28
26
|
tensor(int16),
|
|
29
27
|
tensor(int32),
|
|
30
28
|
tensor(int64),
|
|
31
29
|
tensor(int8),
|
|
32
30
|
tensor(uint16),
|
|
33
31
|
tensor(uint32),
|
|
34
32
|
tensor(uint64),
|
|
35
33
|
tensor(uint8)
|
|
36
34
|
):
|
|
37
35
|
Constrain input Y types to float/int tensors.
|