Dropout - 1 vs 7#

Next section compares an older to a newer version of the same operator after both definition are converted into markdown text.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. Dropout1 → Dropout7 +7 -3
Dropout1 → 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.
7
6
  **Attributes**
7
+ * **consumed_inputs**:
8
+ legacy optimization attribute.
9
+ * **is_test**:
10
+ (int, default 0) if nonzero, run dropout in test mode where the
11
+ output is simply Y = X.
8
12
  * **ratio**:
9
- The ratio of random dropout
13
+ (float, default 0.5) the ratio of random dropout
10
14
  **Inputs**
11
15
  * **data** (heterogeneous) - **T**:
12
16
  The input data as Tensor.
13
17
  **Outputs**
14
18
  Between 1 and 2 outputs.
15
19
  * **output** (heterogeneous) - **T**:
16
20
  The output.
17
21
  * **mask** (optional, heterogeneous) - **T**:
18
- The output mask.
22
+ The output mask. If is_test is nonzero, this output is not filled.
19
23
  **Type Constraints**
20
24
  * **T** in (
21
25
  tensor(double),
22
26
  tensor(float),
23
27
  tensor(float16)
24
28
  ):
25
29
  Constrain input and output types to float tensors.