CumSum - 11 vs 14¶
Next section compares an older to a newer version of the same operator after both definition are converted into markdown text.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.
- CumSum11 → CumSum14 +2 -2
CumSum11 → CumSum14
RENAMED
@@ -1 +1 @@
|
|
1
1
|
Performs cumulative sum of the input elements along the given axis.
|
2
2
|
By default, it will do the sum inclusively meaning the first element is copied as is.
|
3
3
|
Through an exclusive attribute, this behavior can change to exclude the first element.
|
4
4
|
It can also perform summation in the opposite direction of the axis. For that, set reverse attribute to 1.
|
5
5
|
Example:
|
6
6
|
input_x = [1, 2, 3]
|
7
7
|
axis=0
|
8
8
|
output = [1, 3, 6]
|
9
9
|
exclusive=1
|
10
10
|
output = [0, 1, 3]
|
11
11
|
exclusive=0
|
12
12
|
reverse=1
|
13
13
|
output = [6, 5, 3]
|
14
14
|
exclusive=1
|
15
15
|
reverse=1
|
16
16
|
output = [5, 3, 0]
|
17
17
|
### Attributes
|
18
18
|
* **exclusive - INT** (default is '0'):
|
19
19
|
If set to 1 will return exclusive sum in which the top element is not included. In other terms, if set to 1, the j-th output element would be the sum of the first (j-1) elements. Otherwise, it would be the sum of the first j elements.
|
20
20
|
* **reverse - INT** (default is '0'):
|
21
21
|
If set to 1 will perform the sums in reverse direction.
|
22
22
|
### Inputs
|
23
23
|
- **x** (heterogeneous) - **T**:
|
24
24
|
An input tensor that is to be processed.
|
25
25
|
- **axis** (heterogeneous) - **T2**:
|
26
26
|
A 0-D tensor. Must be in the range [-rank(x), rank(x)-1]. Negative value means counting dimensions from the back.
|
27
27
|
### Outputs
|
28
28
|
- **y** (heterogeneous) - **T**:
|
29
29
|
Output tensor of the same type as 'x' with cumulative sums of the x's elements
|
30
30
|
### Type Constraints
|
31
|
-
* **T** in ( tensor(double), tensor(float), tensor(int32), tensor(int64), tensor(uint32), tensor(uint64) ):
|
31
|
+
* **T** in ( tensor(bfloat16), tensor(double), tensor(float), tensor(float16), tensor(int32), tensor(int64), tensor(uint32), tensor(uint64) ):
|
32
|
-
|
32
|
+
Constrain input and output types to high-precision numeric tensors.
|
33
33
|
* **T2** in ( tensor(int32), tensor(int64) ):
|
34
34
|
axis tensor can be int32 or int64 only
|