ReduceSumSquare - 1 vs 18¶
Next section compares an older to a newer version of the same operator after both definition are converted into markdown text.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.
ReduceSumSquare1 → ReduceSumSquare18
RENAMED
@@ -1 +1 @@
|
|
1
|
-
Computes the sum square of the input tensor's
|
1
|
+
Computes the sum square of the input tensor's elements along the provided axes. The resulting
|
2
|
-
tensor has the same rank as the input if keepdims equals 1. If keepdims
|
2
|
+
tensor has the same rank as the input if keepdims equals 1. If keepdims equals 0, then
|
3
|
-
the
|
3
|
+
the resulting tensor has the reduced dimension pruned. Input tensors of rank zero are
|
4
4
|
valid. Reduction over an empty set of values yields 0.
|
5
|
-
The above behavior is similar to numpy, with the exception that numpy defaults keepdims
|
5
|
+
The above behavior is similar to numpy, with the exception that numpy defaults keepdims
|
6
|
-
False instead of True.
|
6
|
+
to False instead of True.
|
7
|
+
|
8
|
+
#### Function Body
|
9
|
+
|
10
|
+
The function definition for this operator.
|
11
|
+
|
12
|
+
|
13
|
+
<
|
14
|
+
domain: "",
|
15
|
+
opset_import: ["" : 18]
|
16
|
+
>
|
17
|
+
ReduceSumSquare <noop_with_empty_axes,keepdims>(data, axes) => (reduced)
|
18
|
+
{
|
19
|
+
data_square = Mul (data, data)
|
20
|
+
reduced = ReduceSum <keepdims: int = @keepdims, noop_with_empty_axes: int = @noop_with_empty_axes> (data_square, axes)
|
21
|
+
}
|
22
|
+
|
7
23
|
### Attributes
|
8
|
-
|
9
|
-
* **axes - INTS** :
|
10
|
-
|
11
|
-
A list of integers, along which to reduce. The default is to reduce over all the dimensions of the input tensor.
|
12
24
|
* **keepdims - INT** (default is '1'):
|
13
25
|
Keep the reduced dimension or not, default 1 means keep reduced dimension.
|
26
|
+
* **noop_with_empty_axes - INT** (default is '0'):
|
27
|
+
|
28
|
+
Defines behavior when axes is not provided or is empty. If false (default), reduction happens over all axes. If true, no reduction is applied, but other operations will be performed. For example, ReduceSumSquare acts as a vanilla Square.
|
29
|
+
|
14
30
|
### Inputs
|
31
|
+
|
32
|
+
Between 1 and 2 inputs.
|
15
33
|
- **data** (heterogeneous) - **T**:
|
16
34
|
An input tensor.
|
35
|
+
- **axes** (optional, heterogeneous) - **tensor(int64)**:
|
36
|
+
|
37
|
+
Optional input list of integers, along which to reduce. The default is to reduce over empty axes. When axes is empty (either not provided or explicitly empty), behavior depends on 'noop_with_empty_axes': reduction over all axes if 'noop_with_empty_axes' is false, or no reduction is applied if 'noop_with_empty_axes' is true (but other operations will be performed). Accepted range is [-r, r-1] where r = rank(data).
|
17
38
|
### Outputs
|
18
39
|
- **reduced** (heterogeneous) - **T**:
|
19
40
|
Reduced output tensor.
|
20
41
|
### Type Constraints
|
21
|
-
* **T** in ( tensor(double), tensor(float), tensor(float16), tensor(int32), tensor(int64), tensor(uint32), tensor(uint64) ):
|
42
|
+
* **T** in ( tensor(bfloat16), tensor(double), tensor(float), tensor(float16), tensor(int32), tensor(int64), tensor(uint32), tensor(uint64) ):
|
22
|
-
Constrain input and output types to
|
43
|
+
Constrain input and output types to numeric tensors.
|