Where - 9 vs 16

Files changed (1) hide show
  1. Where9 → Where16 +6 -1
Where9 → Where16 RENAMED
@@ -1 +1 @@
1
1
  Return elements, either from X or Y, depending on condition.
2
2
  Where behaves like
3
3
  [numpy.where](https://docs.scipy.org/doc/numpy/reference/generated/numpy.where.html)
4
4
  with three parameters.
5
5
  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>_.
6
+
7
+ **History**
8
+ - Version 16 adds bfloat16 to the types allowed (for the second and third parameter).
6
9
  **Inputs**
7
10
  * **condition** (heterogeneous) - **B**:
8
11
  When True (nonzero), yield X, otherwise yield Y
9
12
  * **X** (heterogeneous) - **T**:
10
13
  values selected at indices where condition is True
11
14
  * **Y** (heterogeneous) - **T**:
12
15
  values selected at indices where condition is False
13
16
  **Outputs**
14
17
  * **output** (heterogeneous) - **T**:
15
18
  Tensor of shape equal to the broadcasted shape of condition, X, and
16
19
  Y.
17
20
  **Type Constraints**
18
21
  * **B** in (
19
22
  tensor(bool)
20
23
  ):
21
24
  Constrain to boolean tensors.
22
25
  * **T** in (
26
+ tensor(bfloat16),
23
27
  tensor(bool),
24
28
  tensor(complex128),
25
29
  tensor(complex64),
26
30
  tensor(double),
27
31
  tensor(float),
28
32
  tensor(float16),
29
33
  tensor(int16),
30
34
  tensor(int32),
31
35
  tensor(int64),
32
36
  tensor(int8),
33
37
  tensor(string),
34
38
  tensor(uint16),
35
39
  tensor(uint32),
36
40
  tensor(uint64),
37
41
  tensor(uint8)
38
42
  ):
39
- Constrain input and output types to all tensor types.? ^
43
+ Constrain input and output types to all tensor types (including
44
+ bfloat).