Dropout - 1 vs 6#
Next section compares an older to a newer version of the same operator after both definition are converted into markdown text.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.
- Dropout1 → Dropout6 +2 -0
Dropout1 → Dropout6
RENAMED
|
@@ -1 +1 @@
|
|
|
1
1
|
Dropout takes one input data (Tensor<float>) and produces two Tensor outputs,
|
|
2
2
|
output (Tensor<float>) and mask (Tensor<bool>). Depending on whether it is in
|
|
3
3
|
test mode or not, the output Y will either be a random dropout, or a simple
|
|
4
4
|
copy of the input. Note that our implementation of Dropout does scaling in
|
|
5
5
|
the training phase, so during testing nothing needs to be done.
|
|
6
6
|
**Attributes**
|
|
7
|
+
* **consumed_inputs**:
|
|
8
|
+
legacy optimization attribute.
|
|
7
9
|
* **is_test**:
|
|
8
10
|
(int, default 0) if nonzero, run dropout in test mode where the
|
|
9
11
|
output is simply Y = X.
|
|
10
12
|
* **ratio**:
|
|
11
13
|
(float, default 0.5) the ratio of random dropout
|
|
12
14
|
**Inputs**
|
|
13
15
|
* **data** (heterogeneous) - **T**:
|
|
14
16
|
The input data as Tensor.
|
|
15
17
|
**Outputs**
|
|
16
18
|
Between 1 and 2 outputs.
|
|
17
19
|
* **output** (heterogeneous) - **T**:
|
|
18
20
|
The output.
|
|
19
21
|
* **mask** (optional, heterogeneous) - **T**:
|
|
20
22
|
The output mask. If is_test is nonzero, this output is not filled.
|
|
21
23
|
**Type Constraints**
|
|
22
24
|
* **T** in (
|
|
23
25
|
tensor(double),
|
|
24
26
|
tensor(float),
|
|
25
27
|
tensor(float16)
|
|
26
28
|
):
|
|
27
29
|
Constrain input and output types to float tensors.
|