Tile - 1 vs 6¶
- Tile1 → Tile6 +23 -9
Tile1 → Tile6
RENAMED
@@ -1 +1 @@
|
|
1
|
-
|
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]]
|
2
4
|
**Inputs**
|
3
5
|
* **input** (heterogeneous) - **T**:
|
4
6
|
Input tensor of any shape.
|
5
|
-
* **
|
7
|
+
* **repeats** (heterogeneous) - **T1**:
|
6
|
-
Number of repeated copies to make of the input tensor.
|
7
|
-
* **axis** (heterogeneous) - **T**:
|
8
|
+
1D int64 tensor of the same length as input's dimension number,
|
8
|
-
|
9
|
+
includes numbers of repeated copies along input's dimensions.
|
9
10
|
**Outputs**
|
10
11
|
* **output** (heterogeneous) - **T**:
|
11
|
-
Output tensor of same
|
12
|
+
Output tensor of the same dimensions and type as tensor input.
|
13
|
+
output_dim[i] = input_dim[i] * repeats[i]
|
12
14
|
**Type Constraints**
|
13
15
|
* **T** in (
|
16
|
+
tensor(bool),
|
17
|
+
tensor(complex128),
|
18
|
+
tensor(complex64),
|
14
19
|
tensor(double),
|
15
20
|
tensor(float),
|
16
|
-
tensor(float16)
|
21
|
+
tensor(float16),
|
22
|
+
tensor(int16),
|
23
|
+
tensor(int32),
|
24
|
+
tensor(int64),
|
25
|
+
tensor(int8),
|
26
|
+
tensor(string),
|
27
|
+
tensor(uint16),
|
28
|
+
tensor(uint32),
|
29
|
+
tensor(uint64),
|
30
|
+
tensor(uint8)
|
17
31
|
):
|
18
|
-
Constrain input types to
|
32
|
+
Constrain input and output types to all tensor types.
|
19
33
|
* **T1** in (
|
20
34
|
tensor(int64)
|
21
35
|
):
|
22
|
-
Constrain
|
36
|
+
Constrain repeat's type to int64 tensors.? +++++
|