ReduceSum - 11 vs 13

Files changed (1) hide show
  1. ReduceSum11 → ReduceSum13 +16 -6
ReduceSum11 → ReduceSum13 RENAMED
@@ -1 +1 @@
1
1
  Computes the sum of the input tensor's element along the provided axes. The resulting
2
- tensor has the same rank as the input if keepdims equals 1. If keepdims equal 0, then
2
+ tensor has the same rank as the input if keepdims equals 1. If keepdims equals 0, then
3
- the resulted tensor have the reduced dimension pruned.
3
+ the resulting tensor has the reduced dimension pruned.
4
4
  The above behavior is similar to numpy, with the exception that numpy defaults keepdims to
5
5
  False instead of True.
6
6
  **Attributes**
7
- * **axes**:
8
- A list of integers, along which to reduce. The default is to reduce
9
- over all the dimensions of the input tensor. Accepted range is [-r,
10
- r-1] where r = rank(data).
11
7
  * **keepdims**:
12
8
  Keep the reduced dimension or not, default 1 means keep reduced
13
9
  dimension.
10
+ * **noop_with_empty_axes**:
11
+ Defines behaviour if 'axes' is empty. Default behaviour with 'false'
12
+ is to reduce all axes. When axes is empty and this attribute is set
13
+ to true, input tensor will not be reduced,and the output tensor
14
+ would be equivalent to input tensor.
14
15
  **Inputs**
16
+ Between 1 and 2 inputs.
17
+
15
18
  * **data** (heterogeneous) - **T**:
16
19
  An input tensor.
20
+ * **axes** (optional, heterogeneous) - **tensor(int64)**:
21
+ Optional input list of integers, along which to reduce. The default
22
+ is to reduce over all the dimensions of the input tensor if
23
+ 'noop_with_empty_axes' is false, else act as an Identity op when
24
+ 'noop_with_empty_axes' is true. Accepted range is [-r, r-1] where r
25
+ = rank(data).
17
26
  **Outputs**
18
27
  * **reduced** (heterogeneous) - **T**:
19
28
  Reduced output tensor.
20
29
  **Type Constraints**
21
30
  * **T** in (
31
+ tensor(bfloat16),
22
32
  tensor(double),
23
33
  tensor(float),
24
34
  tensor(float16),
25
35
  tensor(int32),
26
36
  tensor(int64),
27
37
  tensor(uint32),
28
38
  tensor(uint64)
29
39
  ):
30
40
  Constrain input and output types to high-precision numeric tensors.