Dropout - 6 vs 7

Files changed (1) hide show
  1. Dropout6 → Dropout7 +3 -5
Dropout6 → Dropout7 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
+ This operator has **optional** inputs/outputs. See ONNX <https://github.com/onnx/onnx/blob/master/docs/IR.md>_ for more details about the representation of optional arguments. An empty string may be used in the place of an actual argument's name to indicate a missing argument. Trailing optional arguments (those not followed by an argument that is present) may also be simply omitted.
6
7
  **Attributes**
7
- * **is_test**:
8
- (int, default 0) if nonzero, run dropout in test mode where the
9
- output is simply Y = X.
10
8
  * **ratio**:
11
- (float, default 0.5) the ratio of random dropout
9
+ The ratio of random dropout
12
10
  **Inputs**
13
11
  * **data** (heterogeneous) - **T**:
14
12
  The input data as Tensor.
15
13
  **Outputs**
16
14
  Between 1 and 2 outputs.
17
15
  * **output** (heterogeneous) - **T**:
18
16
  The output.
19
17
  * **mask** (optional, heterogeneous) - **T**:
20
- The output mask. If is_test is nonzero, this output is not filled.
18
+ The output mask.
21
19
  **Type Constraints**
22
20
  * **T** in (
23
21
  tensor(double),
24
22
  tensor(float),
25
23
  tensor(float16)
26
24
  ):
27
25
  Constrain input and output types to float tensors.