Flatten - 11 vs 13#

Next section compares an older to a newer version of the same operator after both definition are converted into markdown text.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. Flatten11 → Flatten13 +0 -1
Flatten11 → Flatten13 RENAMED
@@ -1 +1 @@
1
1
  Flattens the input tensor into a 2D matrix. If input tensor has shape
2
2
  (d_0, d_1, ... d_n) then the output will have shape
3
3
  (d_0 X d_1 ... d_(axis-1), d_axis X d_(axis+1) ... X dn).
4
4
  **Attributes**
5
5
  * **axis**:
6
6
  Indicate up to which input dimensions (exclusive) should be
7
7
  flattened to the outer dimension of the output. The value for axis
8
8
  must be in the range [-r, r], where r is the rank of the input
9
9
  tensor. Negative value means counting dimensions from the back. When
10
10
  axis = 0, the shape of the output tensor is (1, (d_0 X d_1 ... d_n),
11
11
  where the shape of the input tensor is (d_0, d_1, ... d_n).
12
12
  **Inputs**
13
13
  * **input** (heterogeneous) - **T**:
14
14
  A tensor of rank >= axis.
15
15
  **Outputs**
16
16
  * **output** (heterogeneous) - **T**:
17
17
  A 2D tensor with the contents of the input tensor, with input
18
18
  dimensions up to axis flattened to the outer dimension of the output
19
19
  and remaining input dimensions flattened into the inner dimension of
20
20
  the output.
21
21
  **Type Constraints**
22
22
  * **T** in (
23
- tensor(bfloat16),
24
23
  tensor(bool),
25
24
  tensor(complex128),
26
25
  tensor(complex64),
27
26
  tensor(double),
28
27
  tensor(float),
29
28
  tensor(float16),
30
29
  tensor(int16),
31
30
  tensor(int32),
32
31
  tensor(int64),
33
32
  tensor(int8),
34
33
  tensor(string),
35
34
  tensor(uint16),
36
35
  tensor(uint32),
37
36
  tensor(uint64),
38
37
  tensor(uint8)
39
38
  ):
40
39
  Constrain input and output to all tensor types.