Unsqueeze - 1 vs 11#

Next section compares an older to a newer version of the same operator after both definition are converted into markdown text.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.

Files changed (1) hide show
  1. Unsqueeze1 → Unsqueeze11 +7 -14
Unsqueeze1 → Unsqueeze11 RENAMED
@@ -1 +1 @@
1
- Insert single-dimensional entries to the shape of an input tensor (data).
1
+ Insert single-dimensional entries to the shape of a tensor.
2
+ Takes one required argument axes, a list of dimensions that will be inserted.
3
+ Dimension indices in axes are as seen in the output tensor. For example:
2
- Takes one required argument axes - which contains a list of dimension indices and this operator will insert a dimension of value 1 into the corresponding index of the output tensor (expanded).
3
-
4
- For example:
5
- Given an input tensor (data) of shape [3, 4, 5], then
4
+ Given a tensor such that tensor with shape [3, 4, 5], then
5
+ Unsqueeze(tensor, axes=[0, 4]) has shape [1, 3, 4, 5, 1]
6
- Unsqueeze(data, axes=[0, 4]) outputs a tensor (expanded) containing same data as data but with shape [1, 3, 4, 5, 1].
7
-
8
- The attribute axes should not contain any duplicate entries. It is an error if it contains duplicates.
9
- The rank of the output tensor (output_rank) is the rank of the input tensor (data) plus the number of values in axes.
10
- Each value in axes should be within the (inclusive) range [-output_rank , output_rank - 1].
11
- The order of values in axes does not matter and can come in any order.
12
6
  **Attributes**
13
7
  * **axes** (required):
8
+ List of non-negative integers, indicate the dimensions to be
9
+ inserted
14
- List of integers indicating the dimensions to be inserted. Negative
15
- value means counting dimensions from the back. Accepted range is
16
- [-r, r-1] where r = rank(expanded).
17
10
  **Inputs**
18
11
  * **data** (heterogeneous) - **T**:
19
12
  Original tensor
20
13
  **Outputs**
21
14
  * **expanded** (heterogeneous) - **T**:
22
15
  Reshaped tensor with same data as input.
23
16
  **Type Constraints**
24
17
  * **T** in (
25
18
  tensor(bool),
26
19
  tensor(complex128),
27
20
  tensor(complex64),
28
21
  tensor(double),
29
22
  tensor(float),
30
23
  tensor(float16),
31
24
  tensor(int16),
32
25
  tensor(int32),
33
26
  tensor(int64),
34
27
  tensor(int8),
35
28
  tensor(string),
36
29
  tensor(uint16),
37
30
  tensor(uint32),
38
31
  tensor(uint64),
39
32
  tensor(uint8)
40
33
  ):
41
34
  Constrain input and output types to all tensor types.