Reshape - 1 vs 5#
Next section compares an older to a newer version of the same operator after both definition are converted into markdown text.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.
- Reshape1 → Reshape5 +10 -17
Reshape1 → Reshape5
RENAMED
@@ -1 +1 @@
|
|
1
1
|
Reshape the input tensor similar to numpy.reshape.
|
2
|
-
|
2
|
+
It takes a tensor as input and an argument shape. It outputs the reshaped tensor.
|
3
3
|
At most one dimension of the new shape can be -1. In this case, the value is
|
4
4
|
inferred from the size of the tensor and the remaining dimensions. A dimension
|
5
5
|
could also be 0, in which case the actual dimension value is unchanged (i.e. taken
|
6
6
|
from the input tensor). Shape (second input) could be an empty shape, which means converting to a scalar.
|
7
7
|
The input tensor's shape and the output tensor's shape are required to have the same number of elements.
|
8
|
+
**Attributes**
|
9
|
+
|
10
|
+
* **consumed_inputs**:
|
11
|
+
legacy optimization attribute.
|
12
|
+
* **shape**:
|
13
|
+
New shape
|
14
|
+
|
8
15
|
**Inputs**
|
9
16
|
* **data** (heterogeneous) - **T**:
|
10
17
|
An input tensor.
|
11
|
-
* **shape** (heterogeneous) - **tensor(int64)**:
|
12
|
-
Specified shape for output.
|
13
18
|
**Outputs**
|
14
19
|
* **reshaped** (heterogeneous) - **T**:
|
15
20
|
Reshaped data.
|
16
21
|
**Type Constraints**
|
17
22
|
* **T** in (
|
18
|
-
tensor(bool),
|
19
|
-
tensor(complex128),
|
20
|
-
tensor(complex64),
|
21
23
|
tensor(double),
|
22
24
|
tensor(float),
|
23
|
-
tensor(float16)
|
25
|
+
tensor(float16)
|
24
|
-
tensor(int16),
|
25
|
-
tensor(int32),
|
26
|
-
tensor(int64),
|
27
|
-
tensor(int8),
|
28
|
-
tensor(string),
|
29
|
-
tensor(uint16),
|
30
|
-
tensor(uint32),
|
31
|
-
tensor(uint64),
|
32
|
-
tensor(uint8)
|
33
26
|
):
|
34
|
-
Constrain input and output types to
|
27
|
+
Constrain input and output types to float tensors.? +++ ^
|