Tile - 1 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Next section compares an older to a newer version of the same operator after both definition are converted into markdown text.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. Tile1 → Tile13 +9 -24
Tile1 → Tile13 RENAMED
@@ -1 +1 @@
1
+ Repeat the elements of a tensor along an axis.
1
- Constructs a tensor by tiling a given tensor.
2
- This is the same as function tile in Numpy, but no broadcast.
3
- For example A = [[1, 2], [3, 4]], B = [1, 2], tile(A, B) = [[1, 2, 1, 2], [3, 4, 3, 4]]
4
2
  **Inputs**
5
3
  * **input** (heterogeneous) - **T**:
6
4
  Input tensor of any shape.
5
+ * **tiles** (heterogeneous) - **T**:
6
+ Number of repeated copies to make of the input tensor.
7
- * **repeats** (heterogeneous) - **T1**:
7
+ * **axis** (heterogeneous) - **T**:
8
+ Axis along which to repeat.
8
- 1D int64 tensor of the same length as input's dimension number,
9
- includes numbers of repeated copies along input's dimensions.
10
9
  **Outputs**
11
10
  * **output** (heterogeneous) - **T**:
12
- Output tensor of the same dimensions and type as tensor input.
11
+ Output tensor of same shape and type as input.
13
- output_dim[i] = input_dim[i] * repeats[i]
14
12
  **Type Constraints**
15
13
  * **T** in (
16
- tensor(bfloat16),
17
- tensor(bool),
18
- tensor(complex128),
19
- tensor(complex64),
20
14
  tensor(double),
21
15
  tensor(float),
22
- tensor(float16),
16
+ tensor(float16)
23
- tensor(int16),
24
- tensor(int32),
25
- tensor(int64),
26
- tensor(int8),
27
- tensor(string),
28
- tensor(uint16),
29
- tensor(uint32),
30
- tensor(uint64),
31
- tensor(uint8)
32
17
  ):
33
- Constrain input and output types to all tensor types.
18
+ Constrain input types to float tensors.
34
19
  * **T1** in (
35
20
  tensor(int64)
36
21
  ):
37
- Constrain repeat's type to int64 tensors.? ^ ^^ ^
22
+ Constrain tiles and axis's type to int64 tensors.? ^^^ ^^ ^^^^^^^